FORWARD

....................................................................................................................... 17

CHAPTER

THE OVERVIEW .................................... 18

INTRODUCTION ......................................................................................................... 18 OTHER RESOURCES................................................................................................... 19 THE COURSE ............................................................................................................. 19 The rationale........................................................................................................ 20 What you will learn.............................................................................................. 20 Why not NT? ........................................................................................................ 22 COURSE MATERIAL .................................................................................................. 22 Textbook............................................................................................................... 23 85321 CD-ROM................................................................................................... 23 85321 Website...................................................................................................... 23 SOLVING PROBLEMS ................................................................................................. 23 COMPUTERS IN THE REAL WORLD ............................................................................ 24 What you think computers are ............................................................................. 24 Some alternatives................................................................................................. 24 AN OVERVIEW OF LINUX .......................................................................................... 26 Booting................................................................................................................. 27 Running................................................................................................................ 28 Shutting down ...................................................................................................... 29 Layers .................................................................................................................. 29 CONCLUSIONS ........................................................................................................... 31

CHAPTER

........................................................................................................ 32

THE WHAT, WHY AND HOW OF SYS ADMIN ........................... 32

INTRODUCTION ......................................................................................................... 32 WHAT SYSTEMS ADMINISTRATORS DO ..................................................................... 32 Why do we need them .......................................................................................... 33 What they do ........................................................................................................ 33 HOME AND THE REAL WORLD .................................................................................. 36 WHAT SYS ADMINS NEED TO KNOW ......................................................................... 36 WHY UNIX? ............................................................................................................ 38 UNIX PAST, PRESENT AND FUTURE .......................................................................... 39 LINUX ....................................................................................................................... 39 THE RELATIONSHIP BETWEEN LINUX AND UNIX...................................................... 40 SOME MORE SYS ADMIN THEORY ............................................................................. 40 DAILY OPERATIONS .................................................................................................. 41 Automate, automate and automate ...................................................................... 41 System monitoring ............................................................................................... 41 HARDWARE AND SOFTWARE ..................................................................................... 42 Evaluation............................................................................................................ 43 Purchase .............................................................................................................. 43

Installation........................................................................................................... 43 Hardware............................................................................................................. 44 ADMINISTRATION AND PLANNING ............................................................................. 45 Documentation..................................................................................................... 45 POLICY ..................................................................................................................... 48 Penalties .............................................................................................................. 48 Types of Policy..................................................................................................... 48 Creating policy .................................................................................................... 49 CODE OF ETHICS ....................................................................................................... 49 SAGE-AU code of ethics ...................................................................................... 49 SAGE-AU code of ethics ...................................................................................... 50 PEOPLE SKILLS .......................................................................................................... 50 Communicating with Users.................................................................................. 51 How not to communicate with users.................................................................... 54 CONCLUSIONS ........................................................................................................... 54

CHAPTER

........................................................................................................ 55

INFORMATION SOURCES AND PROBLEM SOLVING ........ 55

INTRODUCTION ......................................................................................................... 55 OTHER RESOURCES................................................................................................... 55 INFORMATION SOURCES ........................................................................................... 55 PROFESSIONAL ORGANISATIONS ............................................................................... 56 The SAGE groups ................................................................................................ 56 SAGE-AU............................................................................................................. 57 UNIX User groups ............................................................................................... 57 The ACS, ACM and IEEE .................................................................................... 57 BOOKS AND MAGAZINES ........................................................................................... 57 Bibliographies...................................................................................................... 58 OReilly books ...................................................................................................... 58 Magazines ............................................................................................................ 58 INTERNET RESOURCES .............................................................................................. 58 The 85321 Website............................................................................................... 59 How to use the Internet........................................................................................ 59 Software on the Internet....................................................................................... 59 Discussion forums................................................................................................ 60 Just the FAQs....................................................................................................... 61 Google and Deja News ........................................................................................ 61 Mailing lists ......................................................................................................... 61 Other Discussion Forums .................................................................................... 62 Internet based Linux resources............................................................................ 62 PROBLEM SOLVING ................................................................................................... 63 Guidelines for solving problems.......................................................................... 64 Examples of solving problems ............................................................................. 65 CONCLUSIONS ........................................................................................................... 65 REVIEW QUESTIONS ................................................................................................. 65 INTRODUCTION ......................................................................................................... 66

OTHER RESOURCES................................................................................................... 66 WHAT YOU NEED TO LEARN ...................................................................................... 66 INTRODUCTORY UNIX ............................................................................................. 67 Why do I need to know the command line? ......................................................... 68 How do I learn all this stuff? ............................................................................... 68 UNIX Commands are programs .......................................................................... 69 VI............................................................................................................................. 70 An introduction to vi ........................................................................................... 70 UNIX COMMANDS .................................................................................................... 72 Philosophy of UNIX commands........................................................................... 72 UNIX command format........................................................................................ 73 A command for everything................................................................................... 74 ONLINE HELP ............................................................................................................ 74 Using the manual pages....................................................................................... 75 Is there a man page for... ..................................................................................... 75 man page format .................................................................................................. 75 HTML versions of Manual Pages ........................................................................ 76 SOME UNIX COMMANDS .......................................................................................... 76 Identification Commands..................................................................................... 77 Simple commands ................................................................................................ 78 Filters................................................................................................................... 79 uniq ..................................................................................................................... 80 tr ......................................................................................................................... 81 cut ....................................................................................................................... 81 paste ................................................................................................................... 82 grep ..................................................................................................................... 82 wc ......................................................................................................................... 83 GETTING MORE OUT OF FILTERS ................................................................................ 84 CONCLUSIONS ........................................................................................................... 84

CHAPTER

........................................................................................................ 85

THE FILE HIERARCHY ............................................................................. 85

INTRODUCTION ......................................................................................................... 85 Why? .................................................................................................................... 85 THE IMPORTANT SECTIONS ....................................................................................... 86 The root of the problem ....................................................................................... 86 HOMES FOR USERS .................................................................................................... 87 Every user needs a home... .................................................................................. 87 Other homes?....................................................................................................... 88 /USR AND /VAR .................................................................................................... 88 And the difference is... ......................................................................................... 88 /usr/local ..................................................................................................... 89 lib, include and src .............................................................................................. 90 /var/spool ..................................................................................................... 90 X Windows ........................................................................................................... 91 BINS.......................................................................................................................... 91

Which bin? ........................................................................................................... 91 /bin ..................................................................................................................... 92 /sbin ................................................................................................................... 92 /usr/bin ............................................................................................................. 93 /usr/local/bin ................................................................................................. 93 CONFIGURATION FILES, LOGS AND OTHER BITS! ....................................................... 93 etc etc etc. ............................................................................................................ 93 Logs...................................................................................................................... 94 /proc ................................................................................................................... 94 /dev ..................................................................................................................... 94 CONCLUSION ............................................................................................................ 94 Future standards.................................................................................................. 94 REVIEW QUESTIONS ................................................................................................. 95 4.1 ........................................................................................................................ 95 4.2 ........................................................................................................................ 95 4.3 ........................................................................................................................ 95

CHAPTER

PROCESSES AND FILES ........................................... 96

INTRODUCTION ......................................................................................................... 96 OTHER RESOURCES................................................................................................... 96 MULTIPLE USERS ...................................................................................................... 97 Identifying users................................................................................................... 97 Users and groups................................................................................................. 97 Names and numbers............................................................................................. 97 id ......................................................................................................................... 98 COMMANDS AND PROCESSES .................................................................................... 98 Where are the commands?................................................................................... 98 which ................................................................................................................... 98 Why cant I run my shell script? .......................................................................... 99 When is a command not a command? ............................................................... 100 Why shell commands are faster than other commands ..................................... 100 CONTROLLING PROCESSES ...................................................................................... 101 Viewing existing processes ................................................................................ 101 Job control ......................................................................................................... 105 Manipulating processes ..................................................................................... 107 PROCESS ATTRIBUTES ............................................................................................. 109 Parent processes ................................................................................................ 109 Process UID and GID ....................................................................................... 110 Real UID and GID............................................................................................. 110 Effective UID and GID ...................................................................................... 110 FILES ...................................................................................................................... 111 File types............................................................................................................ 112 Types of normal files.......................................................................................... 112 File attributes..................................................................................................... 113 Viewing file attributes........................................................................................ 114 FILE PROTECTION .................................................................................................... 116 File operations................................................................................................... 116

Users, groups and others................................................................................... 116 Three sets of file permissions............................................................................. 117 Special permissions ........................................................................................... 118 EFFECTIVE UID AND GID ...................................................................................... 119 setuid and setgid ................................................................................................ 119 NUMERIC PERMISSIONS ........................................................................................... 120 Symbolic to numeric .......................................................................................... 121 Exercises ............................................................................................................ 121 CHANGING FILE PERMISSIONS ................................................................................. 122 chmod................................................................................................................. 122 chown................................................................................................................. 123 chgrp ................................................................................................................ 123 chown and chgrp................................................................................................ 124 Default permissions ........................................................................................... 125 FILE PERMISSIONS AND DIRECTORIES ...................................................................... 126 For example ....................................................................................................... 126 What happens if? ............................................................................................... 126 LINKS ..................................................................................................................... 127 Creating Links ................................................................................................... 128 Hard and soft links, the differences ................................................................... 129 SEARCHING THE FILE HIERARCHY ........................................................................... 131 The find command............................................................................................ 131 Exercises ............................................................................................................ 136 PERFORMING COMMANDS ON MANY FILES .............................................................. 137 find and -exec ................................................................................................. 137 find and back quotes....................................................................................... 138 find and xargs ................................................................................................. 138 CONCLUSION .......................................................................................................... 139 REVIEW QUESTIONS ............................................................................................... 140

CHAPTER

...................................................................................................... 142

THE SHELL ....................................................................................................... 142

INTRODUCTION ....................................................................................................... 142 EXECUTING COMMANDS ......................................................................................... 142 Different shells................................................................................................... 143 Starting a shell................................................................................................... 143 PARSING THE COMMAND LINE ................................................................................. 144 THE COMMAND LINE .............................................................................................. 145 Arguments .......................................................................................................... 145 One command to a line ...................................................................................... 146 Commands in the background ........................................................................... 146 FILENAME SUBSTITUTION ....................................................................................... 147 Exercises ............................................................................................................ 149 REMOVING SPECIAL MEANING ................................................................................ 149 INPUT/OUTPUT REDIRECTION .................................................................................. 151 How it works ...................................................................................................... 151

File descriptors.................................................................................................. 152 Standard file descriptors.................................................................................... 152 Changing direction ............................................................................................ 152 Using standard I/O ............................................................................................ 153 Filters................................................................................................................. 153 I/O redirection examples ................................................................................... 154 Redirecting standard error ................................................................................ 155 Evaluating from left to right .............................................................................. 155 EVERYTHING IS A FILE ............................................................................................ 157 tty ..................................................................................................................... 157 Device files......................................................................................................... 157 Redirecting I/O to device files ........................................................................... 158 SHELL VARIABLES .................................................................................................. 159 Environment control .......................................................................................... 159 The set command.............................................................................................. 159 USING SHELL VARIABLES ........................................................................................ 159 Assigning a value............................................................................................... 159 Accessing a variables value.............................................................................. 160 Uninitialised variables....................................................................................... 160 Resetting a variable ........................................................................................... 160 The readonly command.................................................................................... 160 The unset command.......................................................................................... 160 Arithmetic .......................................................................................................... 161 The expr command............................................................................................ 161 Alternatives to expr for arithmetic..................................................................... 162 VALID VARIABLE NAMES ........................................................................................ 162 {} ........................................................................................................................ 162 ENVIRONMENT CONTROL ........................................................................................ 163 PS1 and PS2 ....................................................................................................... 163 bash extensions.................................................................................................. 163 VARIABLES AND SUB-SHELLS ................................................................................. 164 For example ....................................................................................................... 164 export ............................................................................................................... 164 Local variables .................................................................................................. 165 ADVANCED VARIABLE SUBSTITUTION ..................................................................... 165 EVALUATION ORDER ............................................................................................... 166 Why order is important...................................................................................... 166 The order ........................................................................................................... 166 THE EVAL COMMAND............................................................................................. 167 Doing it twice..................................................................................................... 167 CONCLUSION .......................................................................................................... 167 REVIEW QUESTIONS ............................................................................................... 168 INTRODUCTION ....................................................................................................... 171 REGULAR EXPRESSIONS .......................................................................................... 171 REs versus filename substitution ....................................................................... 172 How they work ................................................................................................... 174 REPETITION ............................................................................................................ 174 CONCATENATION AND ALTERNATION .................................................................... 176 DIFFERENT COMMANDS, DIFFERENT RES ............................................................... 176 Exercises ............................................................................................................ 176

TAGGING ................................................................................................................ 176 For example ....................................................................................................... 177 Exercises ............................................................................................................ 177 EX, ED, SED AND VI ......................................................................................... 177 So??? ................................................................................................................. 178 Why use ed? ....................................................................................................... 178 ed commands ..................................................................................................... 178 For example ....................................................................................................... 181 The sed command............................................................................................. 182 sed command format ........................................................................................ 182 UNDERSTANDING COMPLEX COMMANDS ................................................................ 183 CONCLUSIONS ......................................................................................................... 185 REVIEW QUESTIONS ............................................................................................... 185

CHAPTER

SHELL PROGRAMMING ........................................... 186

INTRODUCTION ....................................................................................................... 186 Shell Programming - WHY? .............................................................................. 186 Shell Programming - WHAT?............................................................................ 186 Shell Programming - HOW? ............................................................................. 187 THE BASICS ............................................................................................................ 187 A Basic Program................................................................................................ 187 An Explanation of the Program......................................................................... 189 ALL YOU EVER WANTED TO KNOW ABOUT VARIABLES ........................................ 190 Why? .................................................................................................................. 191 Predefined Variables ......................................................................................... 191 Parameters - Special Shell Variables ................................................................ 192 Only Nine Parameters? ..................................................................................... 194 Exercise.............................................................................................................. 195 The difference between $* and $@.................................................................... 195 THE BASICS OF INPUT/OUTPUT (IO)......................................................................... 196 AND NOW FOR THE HARD BITS ................................................................................ 198 Scenario ............................................................................................................. 198 if ... then ... maybe?............................................................................................ 199 Testing Testing................................................................................................... 200 Expressions, expressions! .................................................................................. 202 Exercise.............................................................................................................. 203 All about case .................................................................................................. 203 Loops and Repeated Action Commands ............................................................ 204 while ............................................................................................................... 204 for .................................................................................................................... 205 Problems with running scanit ........................................................................ 206 So what is happening ......................................................................................... 208 Exercises ............................................................................................................ 210 SPEED AND SHELL SCRIPTS...................................................................................... 210 Whats the mistake ............................................................................................. 210 Solution in C ...................................................................................................... 210

Shell solution written by C programmer ........................................................... 211 Shell solution by shell programmer ................................................................... 211 Comparing the solutions.................................................................................... 211 The problem ....................................................................................................... 212 A solution for scanit?......................................................................................... 212 Number of processes.......................................................................................... 212 UNTIL ...................................................................................................................... 213 break and continue .................................................................................... 214 Redirection......................................................................................................... 215 NOW FOR THE REALLY HARD BITS........................................................................... 215 Functional Functions......................................................................................... 215 local ............................................................................................................... 216 The return trip ............................................................................................... 217 DIFFICULT AND NOT COMPULSORY ......................................................................... 217 Recursion: (see "Recursion") ............................................................................ 217 waiting and traping .................................................................................... 218 BUGS AND DEBUGGING .......................................................................................... 222 Method 1 - set.................................................................................................. 223 Method 2 - echo ............................................................................................... 223 Very Common Mistakes ..................................................................................... 223 AND NOW FOR THE REALLY REALLY HARD BITS ...................................................... 224 Writing good shell programs ............................................................................. 224 eval the wonderful! ......................................................................................... 225 STEP-BY-STEP ......................................................................................................... 227 The problem ....................................................................................................... 227 Solving the problem ........................................................................................... 229 The final program - a listing.............................................................................. 238 FINAL NOTES........................................................................................................... 239 REVIEW QUESTIONS ............................................................................................... 240 REFERENCES ........................................................................................................... 241 SOURCE OF SCANIT ................................................................................................. 241

CHAPTER

USERS ..................................................................................... 244

INTRODUCTION ....................................................................................................... 244 OTHER RESOURCES................................................................................................. 244 WHAT IS A UNIX ACCOUNT? ................................................................................. 244 Login names....................................................................................................... 245 Passwords .......................................................................................................... 246 The UID ............................................................................................................. 247 Home directories ............................................................................................... 248 Login shell ......................................................................................................... 248 Dot files.............................................................................................................. 248 Skeleton directories ........................................................................................... 249 The mail file ....................................................................................................... 250 Mail aliases........................................................................................................ 250 ACCOUNT CONFIGURATION FILES ........................................................................... 251

/etc/passwd ..................................................................................................... 252 EVERYONE CAN READ /ETC/PASSWD ...................................................................... 253

This is a problem ............................................................................................... 253 Password matching............................................................................................ 253 The solution ....................................................................................................... 253 Shadow file format............................................................................................. 254 GROUPS .................................................................................................................. 254 /etc/group ....................................................................................................... 254 SPECIAL ACCOUNTS ................................................................................................ 255 root.................................................................................................................. 255 Restricted actions............................................................................................... 256 Be careful........................................................................................................... 256 THE MECHANICS ..................................................................................................... 256 Other considerations ......................................................................................... 256 Pre-requisite Information .................................................................................. 257 Adding an /etc/passwd entry ...................................................................... 257 The initial password .......................................................................................... 258 /etc/group entry .......................................................................................... 258 The home directory ............................................................................................ 258 The startup files ................................................................................................. 258 Setting up mail ................................................................................................... 259 Testing an account............................................................................................. 259 Inform the user................................................................................................... 261 REMOVING AN ACCOUNT ........................................................................................ 261 Disabling an account......................................................................................... 262 THE GOALS OF ACCOUNT CREATION ...................................................................... 262 MAKING IT SIMPLE .................................................................................................. 263 useradd........................................................................................................... 263 userdel and usermod.................................................................................. 263 Graphical Tools ................................................................................................. 263 AUTOMATION ......................................................................................................... 264 Gathering the information ................................................................................. 264 Policy ................................................................................................................. 264 Creating the accounts ........................................................................................ 265 Additional steps ................................................................................................. 265 Changing passwords without interaction .......................................................... 265 DELEGATION .......................................................................................................... 266 ALLOCATING ROOT PRIVILEGE ................................................................................ 266 sudo.................................................................................................................. 266 sudo advantages .............................................................................................. 268 Exercises ............................................................................................................ 268 CONCLUSIONS ......................................................................................................... 268 REVIEW QUESTIONS ............................................................................................... 269

CHAPTER

.............................................................................................. 270

MANAGING FILE SYSTEMS ................................................................. 270

INTRODUCTION ....................................................................................................... 270 What? ................................................................................................................. 270 Why? .................................................................................................................. 270 OTHER RESOURCES................................................................................................. 271 A SCENARIO............................................................................................................ 271 DEVICES - GATEWAYS TO THE KERNEL................................................................... 272 A device is... ....................................................................................................... 272 Device files are... ............................................................................................... 272 Device drivers are.............................................................................................. 272 /dev.................................................................................................................. 272 Physical characteristics of device files.............................................................. 274 Major and minor device numbers are... ............................................................ 275 Finding the devices on your system ................................................................... 275 Why use device files? ......................................................................................... 277 Creating device files .......................................................................................... 278 The use and abuse of device files....................................................................... 279 DEVICES, PARTITIONS AND FILE SYSTEMS .............................................................. 280 Device files and partitions ................................................................................. 280 Partitions and file systems ................................................................................. 281 Partitions and Blocks......................................................................................... 282 Using the partitions ........................................................................................... 282 The Virtual File System ..................................................................................... 283 Dividing up the file hierarchy - why? ................................................................ 284 Scenario Update ................................................................................................ 285 THE LINUX NATIVE FILE SYSTEM - EXT2 ............................................................... 285 Overview ............................................................................................................ 285 I-Nodes............................................................................................................... 285 Physical Structure and Features ....................................................................... 287 CREATING FILE SYSTEMS ........................................................................................ 289 mkfs .................................................................................................................... 289 Scenario Update ................................................................................................ 289 MOUNTING AND UN-MOUNTING PARTITIONS AND DEVICES .................................. 290 Mount................................................................................................................. 290 Mounting with the /etc/fstab file............................................................... 291 Scenario Update ................................................................................................ 292 FILE OPERATIONS ................................................................................................... 293 Creating a file .................................................................................................... 293 Linking files ....................................................................................................... 293 ln....................................................................................................................... 294 CHECKING THE FILE SYSTEM ................................................................................... 295 Why Me? ............................................................................................................ 295 What to do.......................................................................................................... 295 fsck.................................................................................................................. 295 Using fsck ....................................................................................................... 296 What caused the problem?................................................................................. 296 CONCLUSION .......................................................................................................... 297 REVIEW QUESTIONS ................................................................................................ 297

CHAPTER

................................................................................................ 298

BACKUPS ........................................................................................................... 298

INTRODUCTION ....................................................................................................... 298 It isnt just users who accidentally delete files................................................... 298 OTHER RESOURCES................................................................................................. 298 BACKUPS ARENT ENOUGH ...................................................................................... 299 CHARACTERISTICS OF A GOOD BACKUP STRATEGY ................................................. 299 Ease of use ......................................................................................................... 299 Time efficiency ................................................................................................... 300 Ease of restoring files ........................................................................................ 300 Ability to verify backups .................................................................................... 300 Tolerance of faulty media .................................................................................. 300 Portabilty to a range of platforms ..................................................................... 301 CONSIDERATIONS FOR A BACKUP STRATEGY .......................................................... 301 THE COMPONENTS OF BACKUPS .............................................................................. 302 Scheduler ........................................................................................................... 302 Transport ........................................................................................................... 303 Media ................................................................................................................. 304 COMMANDS ............................................................................................................ 304 dump and restore ......................................................................................... 305 USING DUMP AND RESTORE WITHOUT A TAPE............................................................ 307 Our practice file system ..................................................................................... 307 Doing a level 0 dump ......................................................................................... 308 Restoring the backup ......................................................................................... 309 Alternative.......................................................................................................... 309 The tar command............................................................................................. 310 The dd command ............................................................................................... 312 The mt command ............................................................................................... 313 Compression programs...................................................................................... 314 gzip.................................................................................................................. 315 CONCLUSIONS ......................................................................................................... 315 REVIEW QUESTIONS ................................................................................................ 315

CHAPTER

.............................................................................................. 317

STARTUP AND SHUTDOWN ................................................................. 317

INTRODUCTION ....................................................................................................... 317 OTHER RESOURCES................................................................................................. 317 A BOOTING OVERVIEW............................................................................................ 317 FINDING THE KERNEL ............................................................................................. 318 ROM................................................................................................................... 318 The bootstrap program ...................................................................................... 318 BOOTING ON A PC .................................................................................................. 319

On the floppy...................................................................................................... 319 Making a boot disk............................................................................................. 319 Using a boot loader ........................................................................................... 320 STARTING THE KERNEL ........................................................................................... 320 Kernel boot messages ........................................................................................ 321 STARTING THE PROCESSES ...................................................................................... 322 Run levels........................................................................................................... 323 /etc/inittab............................................................................................... 324 SYSTEM CONFIGURATION ....................................................................................... 327 TERMINAL LOGINS .................................................................................................. 328 STARTUP SCRIPTS ................................................................................................... 328 The Linux Process.............................................................................................. 329 WHY WONT IT BOOT? ............................................................................................. 331 Solutions ............................................................................................................ 331 Boot and root disks ............................................................................................ 331 Making a boot and root disk.............................................................................. 332 Using boot and root ........................................................................................... 332 SOLUTIONS TO HARDWARE PROBLEMS .................................................................... 334 Damaged file systems ........................................................................................ 334 Improperly configured kernels .......................................................................... 334 SHUTTING DOWN .................................................................................................... 334 Reasons Shutting down ...................................................................................... 335 Being nice to the users ....................................................................................... 335 COMMANDS TO SHUTDOWN .................................................................................... 336 shutdown ........................................................................................................... 337 What happens..................................................................................................... 337 The other commands.......................................................................................... 338 CONCLUSIONS ......................................................................................................... 338 REVIEW QUESTIONS ............................................................................................... 338

CHAPTER

.............................................................................................. 340

KERNEL .............................................................................................................. 340

THE BIT OF THE NUT THAT YOU EAT? ...................................................................... 340 OTHER RESOURCES................................................................................................. 340 WHY? ..................................................................................................................... 341 HOW? ..................................................................................................................... 342 THE LIFELESS IMAGE............................................................................................... 342 Kernel gizzards .................................................................................................. 343 THE FIRST INCISION................................................................................................. 344 Making the heart beat........................................................................................ 344 Modules.............................................................................................................. 345 THE PROC FILE SYSTEM ........................................................................................... 346 REALLY, WHY BOTHER? ......................................................................................... 348 DOCUMENTATION ................................................................................................... 351 MODIFYING THE KERNEL........................................................................................ 352 COMPILING THE SOURCE ......................................................................................... 353

Configuration..................................................................................................... 355 Dependencies ..................................................................................................... 356 Compilation ....................................................................................................... 356 Common Problems............................................................................................. 358 CONCLUSIONS ......................................................................................................... 359 REVIEW QUESTIONS ............................................................................................... 360

CHAPTER

.............................................................................................. 361

AUTOMATION AND OBSERVATION ............................................. 361

INTRODUCTION ....................................................................................................... 361 OTHER RESOURCES................................................................................................. 361 AUTOMATION AND CRON ......................................................................................... 361 Components of cron .......................................................................................... 362 crontab format.................................................................................................. 362 Creating crontab files ...................................................................................... 364 CURRENT OBSERVATION ........................................................................................ 364 df ....................................................................................................................... 365 du ....................................................................................................................... 365 System Status ..................................................................................................... 366 HISTORICAL OBSERVATION .................................................................................... 369 Managing log and accounting files ................................................................... 370 Centralise........................................................................................................... 370 Security .............................................................................................................. 370 Look at them ...................................................................................................... 371 LOGGING ................................................................................................................ 371 syslog ............................................................................................................... 371 ACCOUNTING .......................................................................................................... 375 Login accounting ............................................................................................... 375 last ................................................................................................................... 375 ac ....................................................................................................................... 376 Process accounting ............................................................................................ 376 So what?............................................................................................................. 377 CONCLUSIONS ......................................................................................................... 378 REVIEW QUESTIONS ............................................................................................... 379

CHAPTER

.............................................................................................. 380

NETWORKS: THE CONNECTION ..................................................... 380

INTRODUCTION ....................................................................................................... 380 OTHER RESOURCES................................................................................................. 380 THE OVERVIEW ...................................................................................................... 381 What you need.................................................................................................... 381 What you do ....................................................................................................... 382

TCP/IP BASICS ....................................................................................................... 382 Hostnames.......................................................................................................... 383 hostname ........................................................................................................ 384 Qualified names ................................................................................................. 384 IP/Internet Addresses......................................................................................... 385 THE INTERNET IS A NETWORK OF NETWORKS.......................................................... 386 Exercises ............................................................................................................ 390 Name resolution................................................................................................. 390 Routing............................................................................................................... 393 Exercises ............................................................................................................ 394 TCP/IP Basics Conclusion ................................................................................ 394 NETWORK HARDWARE ........................................................................................... 395 Network devices ................................................................................................. 395 Ethernet.............................................................................................................. 396 Converting hardware addresses to Internet addresses ..................................... 397 SLIP, PPP and point to point ............................................................................ 398 KERNEL SUPPORT FOR NETWORKING ...................................................................... 399 CONFIGURING THE CONNECTION ............................................................................. 401 The Configuration Process ................................................................................ 401 Configuration Related Tools and Files.............................................................. 402 Configuring the device/interface ....................................................................... 402 Configuring the name resolver .......................................................................... 403 Configuring routing ........................................................................................... 405 NETWORK MANAGEMENT TOOLS ........................................................................ 408 RedHat GUI Networking Tools ......................................................................... 408 nslookup ........................................................................................................... 409 netstat................................................................................................................. 409 traceroute ....................................................................................................... 410 CONCLUSIONS ......................................................................................................... 411 REVIEW QUESTIONS ............................................................................................... 412

CHAPTER

.............................................................................................. 414

NETWORK APPLICATIONS.................................................................. 414

INTRODUCTION ....................................................................................................... 414 OTHER RESOURCES................................................................................................. 414 HOW IT ALL WORKS ................................................................................................ 415 PORTS ..................................................................................................................... 415 Reserved ports ................................................................................................... 416 Look at ports, netstat ................................................................................... 417 NETWORK DAEMONS .............................................................................................. 418 How network daemons start .............................................................................. 418 inetd ................................................................................................................... 419 How it works ...................................................................................................... 420 NETWORK CLIENTS ................................................................................................. 420 The telnet client.................................................................................................. 420 NETWORK PROTOCOLS ............................................................................................ 421

Request for comment (RFCs)............................................................................. 421 Text based protocols .......................................................................................... 422 How it works ...................................................................................................... 422 Exercises ............................................................................................................ 423 SECURITY ............................................................................................................... 424 TCPWrappers/tcpd ............................................................................................ 424 The difference .................................................................................................... 424 WHATS AN INTRANET?........................................................................................... 426 Services on an Intranet ...................................................................................... 426 FILE AND PRINT SHARING ........................................................................................ 427 Samba ................................................................................................................ 427 EMAIL ..................................................................................................................... 429 Email components.............................................................................................. 429 Email Protocols ................................................................................................. 430 Exercises ............................................................................................................ 432 WORLD-WIDE WEB ................................................................................................ 432 CONCLUSIONS ......................................................................................................... 432 REVIEW QUESTIONS ............................................................................................... 433 LOCAL INTRODUCTION ........................................................................................... 434 LINUX SECURITY HOWTO..................................................................................... 434 INTRODUCTION ....................................................................................................... 434 New Versions of this Document......................................................................... 435 Feedback............................................................................................................ 435 Disclaimer.......................................................................................................... 435 Copyright Information ....................................................................................... 435 OVERVIEW .............................................................................................................. 436 Why Do We Need Security? ............................................................................... 436 How Secure Is Secure? ...................................................................................... 436 What Are You Trying to Protect? ...................................................................... 437 Developing A Security Policy ............................................................................ 438 Means of Securing Your Site.............................................................................. 438 Organization of This Document......................................................................... 439 PHYSICAL SECURITY ............................................................................................... 440 Computer locks .................................................................................................. 440 BIOS Security..................................................................................................... 441 Boot Loader Security ......................................................................................... 441 Detecting Physical Security Compromises........................................................ 442 LOCAL SECURITY.................................................................................................... 443 Creating New Accounts ..................................................................................... 443 Root Security...................................................................................................... 443 FILES AND FILESYSTEM SECURITY.......................................................................... 445 Umask Settings................................................................................................... 446 File Permissions ................................................................................................ 447 Integrity Checking with Tripwire Tripwire........................................................ 449 5.4. Trojan Horses ............................................................................................ 450 PASSWORD SECURITY AND ENCRYPTION ................................................................ 450 PGP and Public-Key Cryptography .................................................................. 451 SSL, S-HTTP, HTTPS and S/MIME................................................................... 452 Linux IPSEC Implementations........................................................................... 452 ssh (Secure Shell) and stelnet ........................................................................... 453

PAM - Pluggable Authentication Modules ........................................................ 454 Cryptographic IP Encapsulation (CIPE) .......................................................... 454 Kerberos ............................................................................................................ 455 Shadow Passwords. ........................................................................................... 455 Crack" and "John the Ripper"......................................................................... 456 CFS & TCFS: Cryptographic File Systems....................................................... 456 X11, SVGA and display security........................................................................ 456 KERNEL SECURITY ................................................................................................. 457 2.0 Kernel Compile Options .............................................................................. 457 2.2 Kernel Compile Options .............................................................................. 459 Kernel Devices................................................................................................... 460 NETWORK SECURITY .............................................................................................. 461 Packet Sniffers ................................................................................................... 461 System services and tcp_wrappers .................................................................... 461 Verify Your DNS Information ............................................................................ 462 identd ................................................................................................................. 463 SATAN, ISS, and Other Network Scanners........................................................ 463 Detecting Port Scans ......................................................................................... 464 sendmail , qmail and MTA's ............................................................................. 464 Denial of Service Attacks................................................................................... 465 NFS (Network File System) Security. ................................................................ 466 NIS (Network Information Service) (formerly YP). ........................................... 466 Firewalls ............................................................................................................ 466 IP Chains - Linux Kernel 2.2.x Firewalling ...................................................... 467 VPN's - Virtual Private Networks...................................................................... 467 SECURITY PREPARATION (BEFORE YOU GO ON-LINE).............................................. 468 Make a Full Backup of Your Machine............................................................... 468 Choosing a Good Backup Schedule................................................................... 468 Backup Your RPM or Debian File Database .................................................... 468 Keep Track of Your System Accounting Data.................................................... 469 Apply All New System Updates.......................................................................... 470 WHAT TO DO DURING AND AFTER A BREAKIN ...................................................... 470 Security Compromise Underway. ...................................................................... 470 Security Compromise has already happened .................................................... 471 SECURITY SOURCES ................................................................................................ 473 FTP Sites............................................................................................................ 473 Web Sites............................................................................................................ 473 Mailing Lists ...................................................................................................... 474 Books - Printed Reading Material..................................................................... 474 GLOSSARY .............................................................................................................. 474 FREQUENTLY ASKED QUESTIONS ........................................................................... 475 CONCLUSION .......................................................................................................... 477 ACKNOWLEDGEMENTS ........................................................................................... 477

Forward
The fourth edition of this text sees quite a few updates and additions. Hopefully it also sees the start of a more collaborative process for the development of the text. In recognition of the fact that one person, with other things to do, simply cant maintain a book such as this I am setting up a number of methods which you can contribute to this book via the 85321 website (http://infocom.cqu.edu.au/85321/). Those contributions have already started including Bruce Jamieson, one of the very first 85321 students, was responsible for most of the content in chapters 4, 8, 10 and 13. Bruces original work has since been touched up slightly to keep up with the changing face of Linux. Chapter 17 of the text is now a copy of the Linux Security HOWTO by Kevin Fenzi and Dave Wreski. Janet Jackson has kindly allowed the reuse of one of her articles in the SAGE-AU newsletter in chapter 1. There are also a number of documents which have been quoted and/or referenced throughout the text.

While the above are the direct contributions there are many people who have made indirect contributions to the development of this text including the people behind the LDP and other available Linux documentation, the authors of the many books Ive read over the years and the people who read this text and provide feedback and suggestions. Please, if you find an error (no matter how simple), find an explanation not particularly helpful or have a suggestion for an addition to the text please visit the website and participate. This participation is particularly important given that there are certain to be errors introduced in this edition. The additions were done quickly with a minimum of proof-reading so please report them if you find them.

David Jones http://cq-pan.cqu.edu.au/david-jones/

85321, Systems Administration

Chapter 0: The Overview

Chapter
The Overview

Introduction
Welcome to Chapter 0. Yes, I know it is a strange numbering scheme but this chapter is being added early in 2000 and it is quicker to call it Chapter 0 rather than call it Chapter 1 and then have to renumber all the remaining chapters which have been in the text for a few years. The inclusion of this chapter is due to feedback from previous students in 85321, Systems Administration. It is an attempt to give you an overview of the course and more importantly of computing, Linux and Systems Administration. Many students have commented that they have felt lost in the detail of Linux without any idea of how it all fits together. Hopefully this chapter will go some way towards solving this problem. Hopefully it will provide some sort of small map and compass so you have some idea of where you are and where you are going. I am always keen and willing to hear feedback about this text. If you have useful suggestions please feel free to make them via the various mechanisms which are available on the 85321 website, http://infocom.cqu.edu.au/85321/ This chapter will discuss the following the course A brief overview of 85321 and why it is the way it is. This will also include an introduction to the material we will cover this term. the course material A really quick explanation of how the 85321 CD-ROM, Website, textbook and other material all fits together (at least how I hope it will). solving your problems There is one thing both you and I can be sure of this term. At some stage you will have problems with Linux or 85321. This section will provide some hints and tips on how you should go about solving these problems. Computers Those of you who have not read widely or have experience within the computing industry will think that computing starts and stops with single, stand-alone Windows computers. This couldnt be further from the truth. This section attempts to give you some idea of at least one version of what is out there. Linux Last but not least the chapter provides a quick overview of Linux, how it works and some of the important concepts you will learn about during this course.
Page 18

David Jones (20.01.00)

85321, Systems Administration

Chapter 0: The Overview

Other Resources
All of the chapters in this text will have a section called Other Resources near the start of the chapter. The idea is to point you to other resources which discuss related material. The 85321 website will maintain a more up to date list of resources which will include comments from people about those resources and a space where you can contribute comments and provide pointers to resources you found useful. One of the most common references will be to the Linux Documentation Project (the LDP). The LDP is a collaborative project by many people throughout the Linux community to develop high quality documentation about the Linux system. A mirror of the LDP website is included on the 85321 website/CD-ROM. Other resources which discuss similar material to this chapter includes Online lectures 1, 2 and 3 on the 85321 Website discuss some of the same information covered here. Though some of the information may be a touch old. HOWTOs These are smallish which provide guidance on a particular topic. One HOWTO which covers similar material to this chapter is the UNIX and Internet Fundamentals HOW-TO Guides The LDP also includes a number of guides that are essentially full-blown books (or very close to it). The Linux and Installation and Getting Started Guide contains some good overview material. The Overview of a Linux System from the Linux Systems Administration Guide is also useful. As is the Linux Overview section from the Linux Administration Made Easy Guide (LAME).

The Course
You can get some idea of what 85321, Systems Administration, and to some extent a career as a Systems Administrator from the following poem written by a past 85321 student. Lament of a Linux Student Here I sit broken hearted Loaded X-Windows and can't get it started Off I go in a Tizzy Looks as though tomorrow I'm busy I can guarantee that most students will at some stage be frustrated, annoyed, depressed and entirely sick of 85321, Linux and anyone responsible for it. This can also be said for a career in Systems Administration.

David Jones (20.01.00)

Page 19

85321, Systems Administration

Chapter 0: The Overview

Many of you may have heard of 85321 from other students and others may well have drawn some conclusions based on your previous experiences in courses I have taught. Hopefully many of them (the bad ones hopefully) wont apply. The experience of past students in this course can be summarised as follows, 85321 is enjoyable very practical a lot of work

Hopefully this year you will find the emphasis more on the first two rather than the last one. There have been a lot of changes made to 85321 for the year 2000. Most have been implemented to reduce the amount of work, increase the enjoyment and make sure that someone who passes 85321 actually knows something about Systems Administration and Linux.

The rationale
Why is the course the way it is? There are lots of reasons which contribute but the main ones are You need to learn about Systems Administration. Systems Administration is an essential task, especially given the increasing importance of computers. Systems Administration is difficult. Software and untrained people cant be Systems Administrators. Knowing about Systems Administration will make you a better programmer and computing professional, even if you dont find employment as a Systems Administrator. People only learn by doing. Sure you might be able to recite back to me a whole bunch of facts, commands and concepts and probably even pass an exam. But you wont know how to be a Systems Administrator. To do this you have to experience it.

The last point cannot be emphasised enough. You will learn nothing from this book and course by simply reading about it. You have to get in and get your hands dirty playing around.

What you will learn

The aim of 85321 is to introduce you to the task of Systems Administration, looking after and maintaining complex computer systems. In particular, 85321 aims to produce students who meet the requirements of a Junior Systems Administrator as outlined in the SAGE Job Description booklet (without the 1 or 2 years experience). You can find an excerpt from the Job Description booklet on the 85321 website (http://infocom.cqu.edu.au/85321/). Figure 0.1. provides a graphical representation of the topics introduced in 85321.

David Jones (20.01.00)

Page 20

85321, Systems Administration

Chapter 0: The Overview

An Overview

of

Figure 0.1. the Content of

this Book.

For the first eight chapters of this book we concentrate on the foundations, basic UNIX. You need to become familiar with basic concepts such as UNIX commands, shells, and regular expressions before you can progress to the real Systems Administration topics. All of these foundation concepts will be reused later in the book. The next remaining concepts are covered in chapters 9 through 17 Users and account management People have to be able to use the systems you manage. Chapter 9 examines the issues involved with this on a Linux system. File systems and Backups People use a computer in order to store and manipulate data. That data has to be stored somewhere. Chapters 10 and 11 examine how Linux stores data on hard-drives and how you can perform backups to tape. Startup and Shutdown Operating systems such as Linux and Windows NT are not simple systems. The process to start them up and shut them down is quite complex and problems can arise. Chapter 12 examines the Linux startup and shutdown process. Kernel Many of the services provided by a computer are implemented in the kernel of the operating system. Chapter 13 examines how to configure, compile and install the Linux kernel. Automation and Observation Once your computer is up and running you need to be able to automate tasks and observe what is going on. Chapter 14 examines how to achieve these two tasks on a Linux computer.

David Jones (20.01.00)

Page 21

85321, Systems Administration

Chapter 0: The Overview

Network Without a network connection and network services most modern computers are considered useless. Chapters 15 and 16 examine how to connect, configure and use a Linux computer on a network. Security Ensuring that your computer and its contents is safe from prying eyes is an essential part of any Systems Administrators job. Chapter 17 provides an overview of security on a Linux system.

All these concepts are essential to Systems Administrators regardless of the type of operating system they are using. Someone managing a Windows NT system still needs to ensure the security of the system, connect it to a network, configure it for new drivers and keep an eye out on what is happening.

Why not NT?

A very common question from 85321 students is why are we using Linux? Why arent we using NT. Here are some of my answers to those questions. It is not cheap It costs money to distribute NT to the couple of hundred students doing 85321 in three or four countries. A lot more money than it does to distribute Linux. it is not complete Adding to the cost is that when you get NT you dont get a real Web server, database and a bunch of other important software. it hides its complexity Windows NT is supposed to be easy to administer. After all it is all GUI based. That isnt an argument there are similar GUI based tools for managing UNIX boxes. The problem with GUIs, especially when you are learning about systems, is that GUIs hide how things work. As a Systems Administrator you need to know how things work. You dont need to know that to get it to work you press that button on that dialog box. A trained monkey can work that out. it is closed NT is Microsofts. They own it. They make the rules. If they are unhappy, the change NT. Linux is owned by a community of people who work together to make it better. if you learn Linux you can learn NT Lastly, if you can learn the material in this textbook. Learning how to administer an NT computer is no great difficulty.

Course Material
For 85321 you will (should) have access to this textbook, an 85321 CD-ROM and the 85321 Website. This section gives a brief overview of the relationships between this material.

David Jones (20.01.00)

Page 22

85321, Systems Administration

Chapter 0: The Overview

Textbook
This book provides most of the reading and exercises you will need for 85321. You should end up reading most of it. Electronic copies of the text are available on the 85321 website. You can purchase copies of it from the CQU bookshop. There are a couple of older chapters from this text which are not included with the print version. They are available from the 85321 website/CD-ROM.

85321 CD-ROM
The 85321 CD-ROM will contain a mirror of the 85321 Website. The main use of this CD-ROM is to provide quick access to the website material without requiring you to connect to the Internet. There most useful parts of the 85321 CD-ROM are expected to be the online lectures Recorded early in 1999 these online lectures cover the first 5 or 6 weeks of the semester and include audio. the Linux Documentation Project mirror The LDP is the main source for Linux information. Copies of most of the LDP information is available on the CD-ROM.

The 85321 CD-ROM does not contain a copy of Linux. To install Linux you will need a separate CD. The 85321 CD-ROM is available for purchase from the CQU bookshop.

85321 Website
It is intended that the 85321 website will be the primary site for interaction and information exchange. The 85321 website should always have the most up to date copies of information. The 85321 website will also have a number of features which will enable you to make contributions to improving the site and the unit. Please take the time to visit and become familiar with the website and its features. The URL is http://infocom.cqu.edu.au/85321/

Solving Problems
Students enrolled in 85321 will be nearing the end their degree. It wont be to long until you are computer professionals being employed to do work with computers. When you are a computer professional you will not be able to ask the lecturer how to do something. You will need to know how to solve the problem yourself, to work it out. If there is one thing I hope you learn from 85321 it is the ability to solve your own problems.

David Jones (20.01.00)

Page 23

85321, Systems Administration

Chapter 0: The Overview

Chapter 2 of this textbook offers more details about how you should go about solving problems. Please refer to it.

Computers in the Real World

Chances are most of your experience with computers are with Wintel PCs (computers with Intel CPUs running various versions of the Windows operating system). As with most people your past experience colours your beliefs. Out in the "real world" (a term I will use throughout the book to refer to largish organisations) there is a lot more to computers than Wintel computers with a single monitor, CPU and keyboard. It is hoped that this section will introduce you to some of the differences you can expect to experience when you enter the "real world".

What you think computers are

Chances are you think computers have one monitor, one CPU, some RAM, a keyboard, a printer, a couple of other perhiperals and maybe a network connection. To use the computer you sit down in front of it, Turn it on. As a result the computer finds some boot information on one of the drives. Loads the kernel of the operating system you use, configures the machine and starts up some other software services. Get presented with a GUI interface, i.e. Windows, on the monitor. Do stuff by double clicking on icons and the like. As a result the computer loads programs from file and executes them using your computers CPU and displays the results on the monitor. You might be able to connect to a network drive. The network drive might contain data or maybe some programs which you can run using the CPU of your computer. When you are finished you turn the computer off.

Some alternatives
This isnt the way it is always. There are computers which differ on any one of the above assumptions. The following introduces you to some of them. No heads There are situations where computers can have no monitors at all (often referred to as headless computers). For example the servers produced by Cobalt Networks (http://www.cobalt.com/) which provide file, print and web serving capabilities are essentially blue boxes with a power and network connection. These servers dont have any peripherals connected to them at all. All management is done via the network. Most multimedia developers will work with two monitors. One which has all their "code" and development work, another which shows the end product.
David Jones (20.01.00) Page 24

85321, Systems Administration

Chapter 0: The Overview

No Graphical User Interface (GUI) GUIs (graphical user interfaces) such as Windows are resource hogs. Running a GUI takes more RAM and larger CPUs than running a text-based command line. With Windows NT you have to run the GUI. With UNIX you can chose to run a GUI or a command line. Since UNIX isnt burdened by a GUI a much smaller machine can run Linux and do the same job as a larger WinNT machine. More CPUs Most personal computers have a single CPU. It is not fairly common for largish network servers to have at least 2 or maybe 4 CPUs. The SUN HPC 10000 (http://www.sun.com/servers/hpc/products/hpc10000.html) will support up to 64 CPUs, 64Gb of RAM and over 64 Terabytes of disk space. Clustering technology, such as Beowulf (http://www.beowulf.org/), allows you to connect multiple personal computers together as a network and treat them as a single computer. No disks Managing a large network of computers where users can modify information on the hard-drive can be a lot of work. People make changes, people make mistakes, Systems Administrator must fix mistake. One solution to this is not to allow people to make changes. In some cases the machines dont even have disks. All the information and programs there computer uses comes from the disks of another computer. In the early 90s some Postscript printers actually had more computing power than the personal computers which were sending them print jobs. Loading programs from a disk on another computer and running them on your own computer is common to both diskless workstations and also to most Windows users. It is common in companies to use a large disk connected to a server for central applications. You want to run Word, well you connect to the network drive that contains Word and run it. The CPU in your machine does the work executing Word but loads it from a network disk. Sharing CPUs UNIX is a true multi-user operating system in the sense that someone on another computer can log onto my computer and run programs. In this situation it is the CPU of my computer (hopefully a large server) which runs the program while the new persons computer takes care of handling the input and output. Under UNIX this can be achieved using telnet for text based programs (this feature is available under Windows) and using features of the X Windows system for GUI-based programs. But alas all is not lost. Virtual Network Computing (VNC) http://www.uk.research.att.com/vnc/ is freely available and provides a similar capability for Windows and Mac computers. In fact, it allows any UNIX, Windows, Mac computer to run applications on any UNIX, Windows Mac

David Jones (20.01.00)

Page 25

85321, Systems Administration

Chapter 0: The Overview

computer and have the output appear on the original computer. It even allows the same effect to be achieved via any Web browser which supports Java. Multiple operating systems, one at a time Up until this course most of you will have been using a single operating system on your computer. It is possible to have more than one operating system on the one computer. The standard approach to achieving this is placing each operating system on its own partition and when you first turn the computer on choosing which operating system you want to run, e.g. WinNT, Linux or Win98. With the standard configuration you cannot have more than one operating system running at any one time. Mainly due to the fact that the operating system provides the interface between the hardware and user programs. So each operating system takes over the hardware. Operating systems have not been able to watch Sesame Street to learn how to share. Running programs from one operating system on another Usually you cannot run a Windows NT program on a computer running Linux computer or a Linux program on a computer running Windows NT. However, there are are systems which aim to allow you to achieve this. The most common under Linux is the Wine system (http://www.winehq.com/) which allows you to run Windows binaries under Linux. Multiple operating systems at the same time In some instances you have to have access to more than one operating system. The above three solutions are workable but have their drawbacks. Using a system like VNC means that you need to have more than one computer. Running multiple operating systems, one at a time, means you have to reboot your computer to change operating systems. The WINE approach isnt quite ready for prime-time use. An alternative approach is provided by VMware (http://www.vmware.com/). VMware provides software which supplies a virtual machine on which you can run other operating systems within other operating systems. For example, using VMware for Linux you can run a copy of Windows NT on the VMware virtual machine and then run any Windows application at the same time as running Linux and Linux applications.

An Overview of Linux
The research literature into education is full of discussion about the advantages of mental models provide towards making learning easier. The idea is that you can only really learn or know something if you build a correct model of the concept you are learning about in your mind. In other words, you will find 85321, Systems Administration and Linux much easier if you have a good idea of how Linux and its various components all work. Hopefully the following will aid you in achieving this goal.

David Jones (20.01.00)

Page 26

85321, Systems Administration

Chapter 0: The Overview

To achieve this goal the section will be divided into four sections booting Which describes what happens when a Linux machine boots. "Boots" is just computer jargon for starting up. Chapter 12 of the text explores the boot process in more detail. running Once a Linux computer has booted it enters another phase, the running phase. This section looks at the various parts of the Linux operating system and supporting tools which enable a Linux computer to "run". Shutting down This section completes the circle. At some stage you will want to turn the Linux computer off. This section describes what happens when you shut it down. The layers Much of the content and concepts introduced in this text will fall into one of a number of layers. This section outlines those layers and offers a brief explanation.

Booting
When you turn a Linux computer on the following happens (similar steps happen for Windows NT and other types of computer) it executes instructions contained within some read only memory (ROM) this usually results in the computer looking in a few places on disks for a boot sector the computer loads the boot sector and executes the instructions contained in that boot sector these instructions generally load the Linux kernel the Linux kernel checks the available hardware, attempts to configure it and then starts up two processes, swap and init the init process then starts executing a bunch of shell scripts contained in the /etc/rc.d directory the startup scripts perform various configuration steps and start a number of daemons

At this stage your Linux computer is running. It is usually in one of a number of run levels. At any one of these stages problems can occur which cause problems. For example, a corrupt boot sector will mean it cant be executed, a kernel which is incompatible with the installed hardware will not be able to use that hardware. As a Systems Administrator you will be responsible for diagnosing and fixing these problems.

David Jones (20.01.00)

Page 27

85321, Systems Administration

Chapter 0: The Overview

Running
Once the computer is up and running you need to start using it. You will need some type of interface to issue commands to the computer and see the results. At the most general level Linux has to types of interface text interface A simple command-line interface is provided by a variety of shells which allow you to enter commands via the keyboard. This is the original interface to UNIX. A graphical interface Linux comes with X Windows which provides all the functionality and features required to implement a graphical user interface (GUI).

The GUI provided by X Windows is significantly more flexible than that of Windows or the Mac. The commands or programs you execute are all stored on disk in files. Rather than stick all those files in together they are separated out using directories. File and directory mean much the same sort of thing as document and folder. Chapters 4, 5 and 10 provide more information about how information is stored in files and directories and the commands you can use to manipulate them. There are a large number of functions which are common to a lot of programs. For example, opening a Window in a GUI, print some text and opening a file. Rather than have every program write their own code to do this Linux comes with a large collection of libraries. These libraries are stored in common locations (e.g. the directories lib /usr/lib and others) and are referred to when needed. The operating system provides a number of low level tasks such as memory management, CPU management, device drivers and the like. Programming libraries provide the executable code to perform slightly higher level tasks required by other programs, such as printing to the screen. Services such as logging onto the system, handling network connections or running the startup scripts are performed by daemons. A daemon is simply a program. It gets started up, usually by the startup scripts when the computer starts, and then sits around waiting for some interesting event to occur. When that event occurs it examines the event performs some appropriate task and then goes back to sleep. There are a large number of daemons on a UNIX system and a fair amount of Systems Administration is dealing with the management and configuration of daemons. This could be quite complex. Thankfully all daemons behave much the same. They generally all have a configuration file Under UNIX most configuration files are text-based. This is good because text is easy to edit and manipulate with normal text processing tools. The configuration file essentially tells the daemon what to do, when and sometimes how. The first thing a daemon will do is read its configuration

David Jones (20.01.00)

Page 28

85321, Systems Administration

Chapter 0: The Overview

file. If you change the configuration file you have to tell the daemon to take another look at the configuration file either by stopping and restarting it or sending it a signal (more on these later) An executable program This is the daemon. The thing which is started and then listens for interesting events. A log file As the daemon performs tasks it will normally record what it is doing in a log file. This allows the Systems Administrator to find out what happened. It is also the primary tool for figuring out what went wrong.

Processes, programs in execution, perform all work on a UNIX system. A process is essentially a bunch of operating system data structures, code and data. The data structures for each process keep a track of the identity of the person who ran the process (in the process user id). The process will only be allowed to perform tasks that the process owner has been allocated. There is one person (account), the root account, which can do anything because permissions are not checked for root. The root account is usually only available to the Systems Administrator. The lack of control placed on the root account means that it is not a good idea to be using the root account to perform normal tasks.

Shutting down
You can't simply turn a UNIX, or for that matter any reasonably complex, computer off. You have to let it have time to tidy up and finish various tasks so it can shutdown in a safe state. The shutdown process involves the init daemon, the same daemon involved in the startup process, executing a number of shell scripts to safely close down the services it started during the boot process.

Layers
A computer system, especially one running Linux, can be though of containing a number of different layers including Hardware At the lost levels is the physical equipment that provides the basic functionality required for the system to perform. At various stages you will be adding or removing hardware from your system. Device drivers The next step up are the device drivers that form part of the Linux kernel. These devices drivers are essentially programs that know how to communicate with specific devices. If you want to use a particular piece of hardware you must have an appropriate device driver included with the kernel of your system. Other kernel services The kernel also provides a number of slightly higher-level services that

David Jones (20.01.00)

Page 29

85321, Systems Administration

Chapter 0: The Overview

have little to do with talking directly with hardware. Examples include CPU scheduling, memory management and file systems. If you want your Linux system to be able to read Windows-98 floppy disks then certain services are required to be included with your kernel. Device files Outside, but closely related to device drivers, are device files. Device files provide a standard interface to devices. This common interface is often used by processes from the next two layers to communicate with the device. Device drivers are little more than translators between the hardware device and the Linux device file interface. No device file often means you cant use the device, even if you have an appropriate device driver. Daemons As introduced earlier daemons are processes that start and then wait some event to occur. Daemons are often used to implement system level services such as starting up the Linux system and allowing people to log on to your Linux system. User programs At the top level are the user programs. These user programs make use of the services provided by daemons, device files and other kernel services to perform tasks required by people. Some of these user programs provide the interface people use, e.g. shells and the X-Window system.

Whats the use of all these layers? Why should I bother understanding them? Well it makes it much easier to identify and fix the problem. Working your way up the layers (from hardware up to user programs) can often be a good approach to solving problems. For example, lets assume I have been unable to connect to my Linux computer over the network. How can I identify and solve the problem Hardware First step is to make sure the hardware is working. For example, is the network connector plugged in, is the Linux computer turned on, can I connect to other similarly located computers. Device drivers Does the kernel on the Linux system contain the appropriate device drivers for the network hardware I am using. Kernel services Are the kernel services required to connect to remote computers correctly installed and configured? Are other similar network services working? Device files Do the appropriate devices exist? Have they been configured correctly? Daemons Do the log files of the associated daemons show any errors? Is the required daemon executing?

And so on. Hopefully you get the idea of slowly progressing up the layers enables you to rule out possibilities.
David Jones (20.01.00) Page 30

85321, Systems Administration

Chapter 0: The Overview

Hopefully the remainder of this text will provide you with the information necessary to now which kernel services are associated with which features of Linux.

Conclusions
Computing is a large field with many different tasks implemented with a plethora of approaches. This chapter has provided a small list of some of the possibilities. These arent the only ones and there are sure to be some new ones developed. As a computing professional you need to be aware of the possibilities. You will only learn Systems Administration by doing Systems Administration.

David Jones (20.01.00)

Page 31

85321, Systems Administration

Chapter 1: The What, Why and How of Sys Admin

Chapter
The What, Why and How of Sys Admin
A beginning is the time for taking the most delicate care that the balances are correct. -- Frank Herbet (Dune)

Introduction
Systems Administration is one of the most complex, fulfilling and misunderstood professions within the computing arena. Everybody who uses the computer depends on the Systems Administrator doing their job correctly and efficiently. However the only time users tend to give the Systems Administrator a second thought is when the computer system is not working. A broken computer system implies some fault on the part of the Systems Administrator.. Very few people, including other computing professionals, understand the complexity and the time-consuming nature of Systems Administration. Even fewer people realise the satisfaction and challenge that Systems Administration presents to the practitioner. It is one of the rare computing professions in which the individual can combine every facet of the computing field into one career (including programming). The aim of this chapter is to provide you with some background to Systems Administration so that you have some idea of why you are reading this and what you may learn via this text.

What Systems Administrators do

Systems Administration is an old responsibility gaining new found importance and acceptance as a profession. It has come into existence because of the increasing complexity of modern computer systems and networks and because of the economys increasing reliance on computers. Any decent size business now requires at least one person to keep the computers running happily. If the computers dont work the business suffers. Smaller companies will dont have the size to support a full-time Systems Administrator and will likely share one (usually some form of consultant) amongst a number of other companies. It can be said that Systems Administrators have two basic reasons for being ensuring that the computing system runs correctly and as efficiently as possible, and ensuring that all users can and do use the computing system to carry out their required work in the easiest and most efficient manner.

People who have studied operating systems may remember these two reasons as being similar to the responsibilities of operating systems. You may also
David Jones (20.01.00) Page 32

85321, Systems Administration

Chapter 1: The What, Why and How of Sys Admin

remember from operating systems that these two responsibilities often conflict with one another. Users will want things a specific way which may not be the best for the organisation. For example, Joe Bloggs in accounting may want this program installed, however the organisation may already have a site licence for another program. The System Administrator, with help from policies, documentation a number of other resources, must attempt to balance these two conflicting aims.

Why do we need them

Every year some company, over the last couple of years it is usually Microsoft, announces some new product which is going to make Systems Administrators obsolete. In fact a couple of the network devices mentioned in Chapter 0 rarely need any form of Systems Administration activities, you set them up and they run. The reason for this is that these types of devices are designed to do one job, e.g. Mail/file/print servers, and nothing else. Their purpose is very specific. However, most organisations cannot be that specific about what they want their computers to do and chances are there wont be a computing device that does exactly what the organisation wants. A lot of the need for Systems Administration is to bridge the gap between what people/organisations want to do and what the computers the organisation has can do.

What they do
The real work required to fulfill these aims depends on the characteristics of the particular computing system and the company it belongs to. Factors that affect what a Systems Administrator needs to do come from a number of categories including: users, hardware , support and policy. Users Users, your colleagues and workmates that use computers and networks to perform their tasks contribute directly to the difficulty (or ease) of your task as a Systems Administrator. Some of the characteristics of people that can contribute to your job include: How many users are there? Two hundred users are more difficult to help than two users and also require completely different practices. With two, or even ten/twenty, users it is possible to become well known to them and really get to know their requirements. With two hundred, or in some cases two thousand users, this is simply not possible. The level of the users expertise. This is a combination of the users actual expertise and their perceived expertise. A user who thinks they know a lot (but doesnt really) can often be more trouble than a user who knows nothing and admits it.

David Jones (20.01.00)

Page 33

85321, Systems Administration

Chapter 1: The What, Why and How of Sys Admin

Users who know what they know. Picture it. You are a Systems Administrator at a United States Air Force base. The people using your machines include people who fly million dollar weapons of destruction that have the ability to reduce buildings if not towns to dust. Your users are supremely confident in their ability. What do you do when an arrogant, abusive Colonel contacts you saying he cannot use his computer? What do you say when you solve the problem by telling him he did not have it plugged in? What do you do when you have to do this more than once? It has happened. What are the users trying to do? If the users are scientists doing research on ground breaking network technology you will be performing completely different tasks than if your users are all doing word processing and spreadsheets. Are they responsible or irresponsible? Do the users follow the rules or do they make their own? Do the users like to play with the machines? Being the Systems Administrator in a computing department at a University, where the users are computing students who want to play and see how far they can go is completely different from working in a government department, where the users hate computing and only use them when necessary. Who do the users know? A user, who has a 15-year-old, computer nerd son can often be the cause of problems since the son will tell the parent all sorts of things about computers and what can be done. Very few people have an appreciation of the constraints placed on a Systems Administrator and the computers under their control. Looking after a home PC is completely different to managing a collection of computers at a place of work.

Hardware/Software The computers, software, networks, printers and other peripherals that are at a site also contribute to the type and amount of work a Systems Administrator must perform. Some considerations include: How many, how big and how complex? Once again greater numbers imply more work. Also it may be more work looking after a network of Windows NT machines than a collection of Windows 3.1 computers. Some sites will have supercomputers, which require specialised knowledge. Is there a network? The existence of a network connecting the machines together raises additional problems and further increases the workload of the Systems Administrator.

David Jones (20.01.00)

Page 34

85321, Systems Administration

Chapter 1: The What, Why and How of Sys Admin

Are the computers heterogeneous or homogenous? Is the hardware and software on every machine the same, or is it different. A great variety in hardware and software will make it much more difficult to manage, especially when there are large numbers. The ability to specify a standard for all computers, in both hardware and software, makes the support job orders of magnitude easier.

Support One other area, which makes a difference to the difficulty of a job as a Systems Administrator, is the level of support in the form of other people, time and resources. The support you do (or dont) receive can take many forms including: Are you alone? At some sites there is one administrator who does everything from installing peripherals, fixing computers, doing backups, helping users find the enter key and a range of other tasks. At other sites these tasks are split amongst a range of administrators, operators and technicians. Are you a full time administrator? In some cases the administrator looks after the machines in addition to performing their "real job".

What are the feelings of staff and management towards the Systems Administrators? In many companies the management and staff see Systems Administrators or other computer support people as overhead. This impression of Systems Administrators as an unnecessary expense influences how the users will act. Similar feelings can occur if previous Systems Administrators have been unprofessional or unable to perform their jobs. Policy (Management) As you read through this text you will be introduced to various forms of policy about the use of computers and networks which are needed. These policies define the what, why and how things are done within an organisation. These can be as petty as always using a specific template for letters, memos and faxes through to something as important as whether or not management can order the Systems Administrator to read another employees email.. Official policies are usually the responsibility of management. It is they who should define the rules and the Systems Administrator who puts them into action. Obviously policy shouldnt be made in a complete vacuum without any knowledge of what is possible (but it often is). Additionally these policies should exist and the people using the systems should be aware of them. If this isnt the case you, or the organisation, can be in trouble legally if you wish to enforce a rule (e.g. You cant send pornographic material to the staff mailing list).

David Jones (20.01.00)

Page 35

85321, Systems Administration

Chapter 1: The What, Why and How of Sys Admin

Home and the Real World

Chances are that your only experience with computing is what you have gained maintaining your computer at home or perhaps helping out a few friends. While useful this experience does not prepare you for what computing is like in the real world, especially in a largish organisation. This small section, along with repreated attempts throughout the remaining chapters of this book (see the Computers and the Real World section in Chapter 0), attempts to provide you with some idea of what is involved with computing in the "real world". Some of the differences you will face in the real world include numbers of users Most Systems Administrators will be responsible for looking after organisations with somewhere between 10 up to 1000s of users. Looking after a small number of users who you know is simple. You can let each person do their own thing and the workload wont be too great. However, with 100s of users you have to standards and policies which are kept. Otherwise you will spend all your time trying to remember the differences and be unable to do some real work. 24x7 operation Increasingly organisations are finding that they must have computer systems available 24 hours a day, 7 days a week. Maintaining this sort of availability requires a number of special steps and rules out a lot of practices which are okay when 24x7 operation isnt an issue. As you progress through the text think about what is written. What implications would 24x7 operation have on the concepts you are reading about.

What Sys Admins need to know

The short and sweet answer is that to be a really good Systems Administrator you need to know everything about the entire computer system including the operating system, hardware, software, users, management, network and anything else you can think of that might affect the system in any way. Failing that lofty aim the System Administrator must have the ability to gain this all-encompassing knowledge. The discovery process may include research, trial and error, or begging. The abilities to learn and problem solve may well be the two most important for a Systems Administrator. At some time during their career a Systems Administrator will make use of knowledge from the following (far from exhaustive) list of fields, both computing and non-computing: programming, Systems Administrators have to be able to program. They might have to write scripts that automate regular tasks or a Visual Basic program to help users perform certain tasks.

David Jones (20.01.00)

Page 36

85321, Systems Administration

Chapter 1: The What, Why and How of Sys Admin

hardware maintenance and installation, This may installing new hardware , cleaning old hardware so that it continues to work or diagnosing problems with hardware. documentation, An essential part of Systems Administrations. Not only must you write documentation for the users of your systems so that they know how to do things. You must also write documentation about what it is you are doing and how you are doing it. This documentation will be used by you and your fellow Systems Administrators. testing, Testing is not an ad hoc process where you try a few things. It is an indepth field on its own. Systems Administrators have to have some idea about testing. You cant put together a system for 1000 users without performing some sort of testing. Human Computer Interface, Writing GUI or Web-based applications are a common task for Sys Admins. Both require some sort of idea about HCI issues. networks and computer communication, Networks are an essential part of any computer system these days. You must be aware of the network and data communications. user education, When Office 2000 comes out do you think all the workers in an organisation teach themselves how to use it? Chances are the Systems Administrator will have to perform some form of training. If you are lucky your organisation might have professionals who look after this form of training. If you are really lucky your organisation might recognise the importance of paying for this training. I wouldnt hold my breath. diplomacy, What happens when the second in charge of an organisation tells you that youre an &%^#$# idiot and shouldnt be working here? Scream back, resort to violence, or run away? A Systems Administrator must be a good talker and able to deal with stressful situations. legal issues and contracts, Unlike many University students most organisations pay for their software (and hardware). This usually involves dealing with some form of licence and legal contracts. Familiarity with these can be very helpful. detective work and problem solving, Following the virtual crumbs to find the cause of a problem can be a lot like detective work. management and policy setting, and public relations.

David Jones (20.01.00)

Page 37

85321, Systems Administration

Chapter 1: The What, Why and How of Sys Admin

Reading The Systems Administrators Guild (SAGE, http://www.usenix.org/sage/) is a professional association for Systems Administrators. SAGE has developed a job description booklet that helps describe what Systems Administrators do and what they need to know. A summary of this book is available from the 85321 Web site/CD-ROM under the Resource Materials section for week 1.

This text and the unit 85321 aim to develop Junior Systems Administrators as specified in the SAGE job descriptions booklet, without the 1 to 3 years experience.

Why UNIX?
Some parts of Systems Administration are independent of the type of computer being used, for example handling user complaints and getting on with management. However by necessity there is a great deal of complex platform dependent knowledge that a Systems Administrator must have in order to carry out their job. One train of thought is that it is impossible to gain a full understanding of Systems Administration without having to grapple with the intricacies of a complex computer system. This is something I believe. This text has been written with the Linux operating system (RedHat version 6.1), a version of UNIX that runs on IBM PC clones, in mind. To get the most out of this book you will need access to the root password of a computer running RedHat version 6.1. The reasons for choosing UNIX, and especially Linux, over any of the other available operating systems include UNIX has a long history both in industry and academia. Knowing UNIX is more likely to help your job prospects than hinder them. UNIX/Linux is one of the current industry buzzwords. It is hardware independent. Linux is free A CD with RedHat Linux can be purchased from the CQU bookshop for less than $(AUD)10. Linux runs on a cheap, popular type of computer. A 386 with 16Mb of RAM can provide mail, web, print and file services for up to 25 users. 486 with 32Mb for up to 100 users. Linux provides the operating system and almost all the other software you require to set up a computer system for a small organisation.
Page 38

David Jones (20.01.00)

85321, Systems Administration

Chapter 1: The What, Why and How of Sys Admin

With Windows NT you will have spend a few thousand dollars, on top of what you spend for the operating system, for a database, web server and other necessary software. If you can learn Linux then learning Windows NT is a piece of cake (and uses many of the same ideas).

Just as there are advantages in using UNIX there are also disadvantages. "My Operating System is better than yours" is a religious war that I dont want to discuss here.

UNIX past, present and future

The history of UNIX is an oft-told tale and it is sometimes hard to pick the right version. The story has been told many ways and the following is one version. Being aware of the history can provide you with some insight into why certain things have been done the way they have Unix History These readings are on the 85321 Web site (or CD-ROM) under the Resource Materials section for week 1. At the current point in time it appears that UNIX has ensconced itself into the following market niches server operating system, and Machines running UNIX are acting as file servers and network servers for local area networks (LANs) of smaller client machines (running MS-DOS, Windows, or Macs). workstation operating system. Workstations are nominally powerful computers usually used by a single user. Engineers, scientists and other people who require a lot of computing power generally use them.

Both these roles are being challenged by the arrival of new operating systems like Windows NT. Linux is slowly making inroads into the personal computing environment. I know of a few companies who now use PCs running Linux, X-Windows and Gnome/KDE as the standard desktop.

Linux
This book has been specifically written to centre on the Linux operating system. Linux was chosen because it is a free, complete version of the UNIX operating system that will run on cheap, entry level machines. The following reading provides you with some background into the development of Linux.

David Jones (20.01.00)

Page 39

85321, Systems Administration

Chapter 1: The What, Why and How of Sys Admin

Linux: What is it and a history These readings are available on the 85321 Web site (or CD-ROM) under the Resource Materials section for week 1.

The relationship between Linux and UNIX

Linux is by no means the only version of UNIX currently available. For example you will find at least three other versions of UNIX being used at CQU Solaris The product of Sun (http://www.sun.com/). These are the guys also responsible for Java. Sun are probably the major commercial UNIX vendor today. Tru64 UNIX The product of Compaq Computer who bought out the original manufacturers Digital. HP/UX Produced by Hewlett Packard

While almost all of the specifics covered in this text and the course 85321 are Linux specific the general concepts are applicable to most versions of UNIX. For example the magic file for Linux may be located in /usr/share/magic but on Solaris it may well be in a different location. However, the concept of a magic file still exists on Solaris. The trick is to remember you will become experienced with Linux specific information. While another version of UNIX will be different you should be able to pick it up quite fast.

Some more Sys Admin theory

Systems Administration is not a responsibility specific to the UNIX operating system. Any company that relies on computers must have Systems Administrators. They may not call them Systems Administrators but studies have shown that it is cheaper to have a full time professional maintaining a companys computers than it is to expect the computer users perform the same tasks. Many of the tasks of Systems Administration are not platform specific. For example a recent survey of Systems Administrators found that 37% of an administrators time is spent helping users. This chapter examines some of the important platform independent tasks that a Systems Administrator must perform. Any Sys Admin that ignores these tasks is going to be in trouble very quickly. For the purposes of this chapter these tasks have been divided up into four categories

David Jones (20.01.00)

Page 40

85321, Systems Administration

Chapter 1: The What, Why and How of Sys Admin

daily operations, hardware and software, interacting with people, and administration and planning.

Daily operations
There are a number of tasks that must be done each day. Some of these tasks are in response to unexpected events, a new user or a system crash, while others are just standard tasks that must be performed regularly.

Automate, automate and automate

A priority for a Systems Administrator must be to automate any task that will be performed regularly. Initially automation may take some additional time, effort and resources but in the long run it will pay off. The benefits of automation include no need to reinvent the wheel, Everytime you need to perform the task you dont have to remember how to do it. it is much simpler, it can be delegated, If the task is simple it can be delegated to someone with less responsibility or it can be completely automated by using the scheduling capabilities of cron (introduced in a later chapter).

For example Obvious examples for automation include adding and removing users, performing backups, and checking disk usage.

System monitoring
This responsibility entails keeping an eye on the state of the computers, software and network to ensure everything is working efficiently. Characteristics of the computer and the operating system that you might keep an eye include resource usage, what people are doing, whether or not the machines normal operations are working.

David Jones (20.01.00)

Page 41

85321, Systems Administration

Chapter 1: The What, Why and How of Sys Admin

Resource usage The operating system and the computer have a number of different resources including disk space, the CPU, RAM, printers and a network. One indication of problems is if anyone person or process is hogging one of these resources. Resource hogging might be an indication of an attack. Steps that might be taken include killing the process that is hogging the resource, changing the process priorities, getting more of the required resource.

What are people doing? As the Systems Administrator you should be aware of what is normal for your site. If the managing director only ever connects between 9 to 5 and his account is currently logged in at 1 in the morning then chances are there is something wrong. Its important not only to observe when but what the users are doing. If the secretary is all of a sudden using the C compiler then theres a good chance that it might not be the secretary. Normal operations Inevitably there will be problems with your system. A disk controller might die, a user might start a run away process that uses all the CPU time, and a mail bounce might result in the hard-drive filling up or any one of millions of other problems. Some of these problems will adversely effect your users. Users will respect you more if they dont have to tell you about problems. Therefore it is important that you maintain a watch on the more important services offered by your computers. You should be watching the services that the users use. Statistics about network, CPU and disk usage are no good when the problem is that the users cant send email because of a problem in the mail configuration. You need to make sure that the users can do what they normally do.

Hardware and software

Major tasks that must be performed with both hardware and software include evaluation, Examining different packages and deciding which is the best for your companys purpose. purchase, Actually obtaining the software, spending the money and signing the contracts.

David Jones (20.01.00)

Page 42

85321, Systems Administration

Chapter 1: The What, Why and How of Sys Admin

installation, Placing the hardware or software onto your system and making it available to the appropriate users. testing and maintenance, Making sure the equipment works and keeping it working. upgrading, Modifying the product to a later version. phasing out. Removing the product from use at your company.

At many companies the Systems Administrator may not have significant say in the evaluation and purchase of a piece of hardware or software. This causes problems because hardware or software is purchased without any consideration of how it will work with existing hardware and software.

Evaluation
Its very hard to convince a software vendor to allow you to return a software package that youve opened, used but found to be unsuitable. The prospect of you making a copy means that most software includes a clause that once you open a packet you own the software and your money wont be refunded. However most vendors recognise the need to evaluate software and supply evaluation versions. These evaluation versions either are a stripped down version with some features turned off, or contain time bomb that makes the package useless after a set date.

Purchase
Under UNIX there are basically two types of software commercial software, or shareware, public domain or free software. Commercial UNIX software will come with the standard agreements and may also include a user limit. The software might be able to be used by 4 or 5 users simultaneously. Most commercial software is managed by licensing software that controls how many copies are being used. As part of the purchase you will receive license numbers that govern how the software may be used. It must be remembered that free software is never free. It still requires time to install, maintain and train users. All this can add up. Some free software can be incredibly easy to install and maintain.

Installation
Most sites will have a policy that covers how and where software must be installed. Some platforms also have software that makes the installation procedure much simpler. It is a very good idea to keep local software separate from the operating system distribution. Mixing them up leads to problems in future upgrades.

David Jones (20.01.00)

Page 43

85321, Systems Administration

Chapter 1: The What, Why and How of Sys Admin

Under Linux and many other modern Unices it is common practice to install all software added locally under the directory /usr/local. There will be more on software installation in a later chapter.

Hardware
At some sites you may have technicians that handle most of the hardware problems. At some sites the Systems Administrator may have to everything from preparing and laying cable through to fixing the fax machine. Either way a Systems Administrator should be capable of performing simple hardware related tasks like installing hard drive and various expansion cards. This isnt the subject to examine hardware related tasks in detail. The following however does provide some simple advice that you should keep in mind. Static electricity Whenever you are handling electrical components you must be aware of static electricity. Static can damage electrical parts. Whenever handling such parts you should be grounded. This is usually achieved by using a static strap. You should be grounded not only when you are installing the parts but at anytime you are handling them. Some people eagerly open packages containing these parts without being grounded. Powering down and wiggling Many hardware faults can be fixed by turning the system off (powering down) and either pushing on the offending card or SIMM (wiggling). Sometimes connectors get dirty and problems can be fixed by cleaning the contacts with a soft pencil eraser (in good condition). Prevention Regular maintenance and prevention tasks can significantly reduce the workload for a Systems Administrator. Some of the common prevention tasks may include ensuring that equipment has a clean, stable power supply, Using power conditioners or uninterruptable power supplies (UPS) to prevent power spikes damaging equipment. ensuring equipment is operating at appropriate temperatures, Make sure that the power vents are clean and unblocked and that air can actually circulate through the equipment. some equipment will need regular lubrication or cleaning, making sure that printers are clean and have sufficient toner, ink etc.

David Jones (20.01.00)

Page 44

85321, Systems Administration

Chapter 1: The What, Why and How of Sys Admin

Administration and planning

This is a task that often receives less attention than others. However it is an essential task that can critically effect your performance as a Systems Administrator. One of the most important aims for a Systems Administrator is to be pro-active rather than reactive. Its very hard for your users to respect you if you are forever badly organised and show no planning ability. Important components of administration and planning include documentation, Both for yourself, the users and management. time management, This is an essential ability for a Systems Administrator who must balance a small amount of time between a large number of simultaneous tasks. policy, There must be policy on just about everything at a site. Having policies that have been accepted by management and hopefully the users is essential. self-education, Computing is always changing. A Systems Administrator must keep up with the pack. planning, What are the aims for your site and yourself for the next 12 months? What major events will happen for which you must prepare? automation, and Anything that can be should be automated. It makes your job easier and gives you more time to do other things. financial planning and management.

Documentation
Documentation is the task that most computing people hate the most and yet is one of the most important tasks for a Systems Administrator. In this context documentation is more than just documentation for users showing them how to use the system. It includes keeping a log book that records all changes made to the system, keeping records and maps of equipment, their location, purchase details etc, Where all the cables are in your building. Which cables connect where. Where are all the machines physically located. labelling hardware, When you are performing maintenance on computers you will need to know information like the type of hard drive controller, number and size of disks, how they are partitioned, hostnames, IP addresses, names of peripherals, any special key strokes or commands for the machine (e.g.

David Jones (20.01.00)

Page 45

85321, Systems Administration

Chapter 1: The What, Why and How of Sys Admin

how to reset the computer) and a variety of other information. Having this information actually on the machine can make maintenance much easier. producing reports, Producing reports of what you are doing and the functioning of the machines is extremely important and will be discussed in more detail later. taking minutes at meetings, and Chances are you will have to attend meetings. Organising, running and recording the minutes of a meeting are all essential skills. producing documentation on how to use the systems at your site. The standard type of documentation required by both users and other Systems Administrators.

Why keep records? It is not unusual for a Systems Administrator to spend two to three days trying to fix some problem that requires minor changes to obscure files hidden away in the dim, dark recesses of the file hierarchy. It is not unusual for a problem of this sort to crop up unexpectedly every six to twelve months. What happens if the Systems Administrator didnt record the solution? Unless he or she is blessed with a photographic memory there is liable to be another two to three days lost trying to fix the problem. Records of everything done to the system must be kept and they must be accessible at all times. What type of records? It is typical for a Systems Administrator and/or a computer site to maintain some type of logbook. There is no set format to follow in keeping a logbook. There are two basic types of logbooks that are used. electronic, or Log information is stored using some type of program or by simply creating a file. paper based. Some form of book or folder in which entries are written by hand. Paper Against

Table 1.1. compares these two forms of logbook. Electronic For easy to update and search easy to include command output Against if the machine is down there is no access to the log can be hard to include diagrams
Electronic

For

less prone to machine harder to update and down time search can be carried around can become messy and hard to read
log books

Table 1.1. versus paper

David Jones (20.01.00)

Page 46

85321, Systems Administration

Chapter 1: The What, Why and How of Sys Admin

What to record? Anything that might be necessary to reconstruct the current state of the computing system should be stored. Examples of necessary information might include copies of policy regarding usernames, directory structure etc, Your site might have a set way of assigning usernames or particular locations in which certain types of files need to be placed. diagrams of the physical connections and layout of the machines and network, Any information required to reconstruct your system, for example CMOS settings on an IBM PC. a copy of a full listing of the device directory, The /dev directory is likely to contain information specific to your machine. If this directory is trashed having a copy in your logbook will enable you to reconstruct it. copies of major configuration files, daily modifications to configuration or other files, lists of useful commands, and solutions to common problems.

Example Log Book Layout The type of information recorded will depend on your responsibilities and the capabilities of your site. There might be someone else who looks after the physical layout of the network leaving you to worry about your machine. It is possible that a logbook might be divided into separate sections. The sections might include configuration information, Listings of the device directory, maps of network and cabling information, and any other static information about the system policy and procedure, A section describing the policy and procedures of the particular machine (usernames, directory locations etc). useful commands, and A list of commands or hints that youve come across that are useful and you would like to remember. daily modifications. The daily modifications made to the system in the normal course of events. The main reason to store this information is so that you have a record of what is being done to your system.

Each entry in a logbook should contain information about time, date, reason for the change, and who made the change.

David Jones (20.01.00)

Page 47

85321, Systems Administration

Chapter 1: The What, Why and How of Sys Admin

If you intend using a paper based logbook then one suggestion is to use a ring binder. Using a ring binder you can add pages to various sections if they start to fill up.

Policy
Think of the computer systems you manage as an environment in which humans live and work. Like any environment, if anarchy is not to reign supreme then there must exist some type of behavioural code that everyone lives by. In a computer system this code is liable to include such things as a single person shall not hog all the resources (disk, cpu etc), users who work for accounting have xyz access, those who work for research have zyx access, and no-one should endeavour to access areas in which they are not allowed.

Penalties
A set of rules by themselves is not enough. There must also exist a set of penalties to be applied if one of the policies is broken, a person(s) charged with detecting the breaking of policy, a person(s) charged with deciding the appropriate policy, a mechanism for the change of policy and penalties, and a mechanism for informing users of the policy and the penalties.

If any one of these necessary components is missing the system may not work to the best of its ability. It is essential that every computer site have widely recognised and accepted policies. The existence of policies ensure consistent treatment of all cases. Policies provide guidelines of what to do in particular cases and what to do if the policies are broken.

Types of Policy
The types of policies you might want to have include the level of service you provide, What operating systems, software etc that you can and will support. What services you provide your users. When will the Systems Administrators or help desk available. the rights and responsibilities of the users, and What they can and cant do. What happens if they break those rules. the rights and responsibilities of the administrators. An often over looked policy. Should Systems Administrators look at other peoples mail?

David Jones (20.01.00)

Page 48

85321, Systems Administration

Chapter 1: The What, Why and How of Sys Admin

Creating policy
Creating policy should include many of the following steps examination of what other similar sites have in the way of policy, widespread involvement of users, management and Systems Administrators in the development of policy, acceptance of policy by management, and checking of the policy by lawyers.

Code of ethics
As the Systems Administrator on a UNIX system you have total control and freedom. All Systems Administrators should follow some form of ethical conduct. The following is a copy of the SAGE-AU Code of Ethical Conduct. The original version is available on the Web at http://www.sageau.org.au/ethics.html.

SAGE-AU code of ethics

In a very short period of time computers have become fundamental to the organisation of societies world-wide; they are now entrenched at every level of human communication from government to the most personal. Computer systems today are not simply constructions of hardware -- rather, they are generated out of an intricate interrelationship between administrators, users, employers, other network sites, and the providers of software, hardware, and national and international communication networks. The demands upon the people who administer these complex systems are wide-ranging. As members of that community of computer managers, and of the System Administrators Guild of Australia (SAGE-AU), we have compiled a set of principles to clarify some of the ethical obligations and responsibilities undertaken by practitioners of this newly emergent profession. We intend that this code will emphasise, both to others and to ourselves, that we are professionals who are resolved to uphold our ethical ideals and obligations. We are committed to maintaining the confidentiality and integrity of the computer systems we manage, for the benefit of all of those involved with them. No single set of rules could apply to the enormous variety of situations and responsibilities that exist: while system administrators must always be guided by their own professional judgment, we hope that consideration of this code will help when difficulties arise. (In this document, the term "users" refers to all people with authorised access to a computer system, including those such as employers, clients, and system staff.)

David Jones (20.01.00)

Page 49

85321, Systems Administration

Chapter 1: The What, Why and How of Sys Admin

SAGE-AU code of ethics

As a member of SAGE-AU I will be guided by the following principles:
Fair Treatment I will treat everyone fairly. I will not discriminate against anyone on grounds such as age, disability, gender, sexual orientation, religion, race, or national origin. Privacy I will access private information on computer systems only when it is necessary in the course of my duties. I will maintain the confidentiality of any information to which I may have access. I acknowledge statutory laws governing data privacy such as the Commonwealth Information Privacy Principles. Communication I will keep users informed about computing matters that may affect them -- such as conditions of acceptable use, sharing of common resources, maintenance of security, occurrence of system monitoring, and any relevant legal obligations. System Integrity I will strive to ensure the integrity of the systems for which I have responsibility, using all appropriate means -- such as regularly maintaining software and hardware; analysing levels of system performance and activity; and, as far as possible, preventing unauthorised use or access. Cooperation I will cooperate with and support my fellow computing professionals. I acknowledge the community responsibility that is fundamental to the integrity of local, national, and international network resources. Honesty I will be honest about my competence and will seek help when necessary. When my professional advice is sought, I will be impartial. I will avoid conflicts of interest; if they do arise I will declare them. Education I will continue to update and enhance my technical knowledge and management skills by training, study, and the sharing of information and experiences with my fellow professionals. Social Responsibility I will continue to enlarge my understanding of the social and legal issues that arise in computing environments, and I will communicate that understanding to others when appropriate. I will strive to ensure that policies and laws about computer systems are consistent with my ethical principles. Workplace Quality I will strive to achieve and maintain a safe, healthy, productive workplace for all users.

People skills
The ability to interact with people is an essential skill for Systems Administrators. The type of people the Systems Administrator must deal with includes users, management, other Systems Administrators and a variety of other people.

David Jones (20.01.00)

Page 50

85321, Systems Administration

Chapter 1: The What, Why and How of Sys Admin

The following reading was first published in "The Australian Systems Administrator" (Vol 1, Issue 2, June/July 1994) the bimonthly newsletter of the Systems Administrators Guild of Australia (SAGE-AU). It provides an example of how a real-life System Administrator handles user liaison.

Communicating with Users

Copyright Janet Jackson Next to balancing conflicting demands, communicating with users is the hardest part of my job. I tend to make a great effort for little gain, whereas in technical endeavours a little effort can produce a major, long-lasting improvement (for example, taking ten minutes to set up regular, automated scratch area cleanups has saved me hours of tedious work and the users a lot of frustration). Also, with users there are emotions to take into account. It doesnt matter whether the computer respects you, but if the users respect you life is a lot easier. My aim in communicating with users is to make life (my job and those of the users) easier by: getting them to respect me (my judgment; my abilities; my integrity and professionalism). teaching them all sorts of things, such as how to remove jobs from the printer queue; what they have to do to keep the systems secure; and when not to interrupt me with questions. In this column Im going to describe some of the communication vehicles Ive tried, and how effective theyve been for me. Ill start with those Ive found least effective overall, and work my way up. Probably the method most useless with the general user community is the policy statement. The typical user just isnt going to read it. However, it can be a good way of communicating with management. Drafting a good policy statement (based on discussions with everyone, but especially with them) shows you mean business and understand how your work fits into the organisation. It should cover the responsibilities of the systems administrator as well as those of the users. Group meetings, whether of the users in general or of a committee of representatives, can help people -- again, especially senior people -- feel more confident that things are going OK, but arent much use for disseminating information. If a meeting is run well you can have a productive discussion of major issues, but if run badly it is likely to turn into a gripe session. Paper memos are to be avoided, because they encourage stiffness and formality. I use them only to answer other peoples paper memos (which are usually complaints) and then only when I dont think the person will read it if I do it by email. Replying by email to a memo has the effect of saying "Theres no need to be so formal". There are a number of leading-the-horse-to-water methods, which only work if the user makes an effort. You can use electronic information services, such as bulletin boards, newsgroups, Gopher, or online manuals; and you can get together a library of printed manuals and books. If you provide easy access to
David Jones (20.01.00) Page 51

85321, Systems Administration

Chapter 1: The What, Why and How of Sys Admin

high-quality information, the interested user can learn a lot. Unfortunately its often the disinterested user that you really want to reach. People often come to my office to ask me things. Youd think that face-to-face communication would work the best, but in this particular setting it doesnt because I am not comfortable. Its not so much that I resent interruptions -- its that I dont have an office, only a desk. Theres no room for a visitors chair; to talk to anyone I have to swivel round and face backwards; and people make a habit of sneaking up on me. Hopefully, one day my campaign for proper accommodation will be successful, and it will be interesting to see how much difference it makes. Talking on the phone is only good for emergencies. Someone is always interrupted; theres no body language; and you tend to forget half of what you wanted to say. I write a column, "Computer Corner", in our staff newsletter. I sometimes write about issues (such as what Im trying to achieve) and sometimes about technical tips. This column isnt as useful as Id hoped. The first problem is that there isnt room to say much, because the newsletter is short and a bit, shall we say, irregular. The second problem is that the rest of the newsletter tends to be kind of dull (lists of visitors; dry field-trip reports; the occasional births and deaths) so people arent so eager to read it. When I pointed this out I was told that it is deliberately impersonal and non-funloving because some of the more senior readers are rather easily offended. Sigh. Next on the scale are signs (on doors, noticeboards, etc) and electronic messages-of-the-day. People have a strong tendency to miss the former and ignore the latter. It may help to make them more interesting with graphics, pictures and human-interest items. Seminars and workshops are worthwhile if you can get people to attend, but theyre a lot of work. If not many turn up, you dont get much return on your investment. Students can sometimes be induced to attend by making it count towards their marks. In other situations, offering food, door prizes, alcohol, sex, drugs or rock-n-roll may help. For explaining specific information (how to pick a good password; how UNIX file permissions work) Ive found paper handouts reasonably effective. Some users take them quite seriously, even filing them for later reference. Unfortunately, others toss them straight in the bin. After about 3 months in my current job I emailed everyone a questionnaire, asking such things as what they used the systems for, what new services they would like to see, and how often they did backups. I offered a chocolate frog to each person who replied. The subject line "Apply here for your FREE chocolate frog" caused some of the more pokerfaced members of staff to delete the mail without reading it, but otherwise the response was surprisingly good. In hindsight, I guess the questionnaire generated more PR than information, although it did confirm my suspicion that most people did not back up their data even though they were supposed to. For me, the second most effective communication vehicle is email. Email is as informal as a personal visit or phone call, but you can get in a lot more information. It is also asynchronous: no-one has to be interrupted, and you dont have to wait for people to be available.

David Jones (20.01.00)

Page 52

85321, Systems Administration

Chapter 1: The What, Why and How of Sys Admin

I often use email broadcasts for notification -- to tell people about impending downtime, for example. Email is quick, convenient, and reaches people who are working offsite. It is also informal and I think people feel more at ease with it than they do with paper memos and printed signs. 1-to-1 email gives people a sense of personal service without much of the hassle that normally entails. At my site people can email problem reports and questions to a special address, "computerhelp". Our stated aim is to respond within 2 working days. We dont always make it. But it does give people a point of contact at all times, even after hours, and it means we get a few less interruptions. Youd think all of that might be enough, but no. My boss said, "You need to communicate more with the users, to tell them about what youre doing". I agreed with him. So I now produce a fortnightly emailed bulletin. It is longer and more formal than a typical email message, with headings and a table of contents. Most of the information in it is positive -- new software that weve installed, and updates on our program of systems improvements. I also include a brief greeting and a couple of witty quotations. Judging by the feedback Ive received, this seems to be working remarkably well -- much better than the staff newsletter column. The only thing that works better than email is personal visits where I am in their office, usually leaning over their screen showing them how to do something. Taking an interest in their work helps a lot. I find this easy where they are graphing the temperature of a lake in glorious colour, but more difficult where they are typing up letters. I dont do enough personal visiting, partly because Im so busy and partly because Im not keen on interrupting people. It usually happens only when theyve asked a question that requires a "show me" approach. A disadvantage of personal visits is that they help only one person at once, whereas with email you can reach all your users. To sum up: in communicating with users, I aim to teach them things and get them to respect me. By sending email I can help the most people for the least effort, although personal visits have much more impact. There are other useful methods, such as policy statements, newsletters, handouts and seminars, but they may not reach the ones who need it most. Its hard. Very hard. If you have any insights or ideas in this area, Id love to hear them, and Im sure the rest of the readers would too. Communicating with management Relationships between Systems Administrators and management can be tense generally because both sides dont understand the importance and problems of the other. Having good Systems Administrators is essential. As is having good management. Management is a difficult task which you wont understand or agree with until you have to perform it. As a Systems Administrator you should keep in mind that the aims of management will not be the same as yours. Management is about profit. When you deal with management keep this in mind. If you need an upgrade of a machine dont argue it on the basis that the load average is running at 5 and the disks are full. Argue it on the basis that due to
David Jones (20.01.00) Page 53

85321, Systems Administration

Chapter 1: The What, Why and How of Sys Admin

the lack of resources the sales force cant take orders and the secretaries are loosing documents which is leading to loss of customers. Generally Systems Administrators tend to focus on achieving a good technical solution. This must be balanced with helping the company you are working for make money.

How not to communicate with users

The Bastard Operator from Hell is a classic (amongst Systems Administrators) collection of stories about a mythically terrible operator. It provides an extreme view of a bad system support person and is also quite funny (depending on your sense of humour). Some of the language may offend some people. Bastard Operator from Hell Available on the 85321 Web site under the Resource Materials section for week 1.

Conclusions
Systems Administration is a complex and interesting field requiring knowledge from most of the fields in computing. It provides a challenging and interesting career. The Linux operating system is an important and available alternative in the current operating systems marked and forms the practical component for this course.

David Jones (20.01.00)

Page 54

85321, Systems Administration

Chapter 2: Information Sources for Sys Admin

Chapter
Information Sources and Problem Solving

Introduction
As a Systems Administrator you will be expected to fix any and all problems that occur with the computer systems under your control. There is no why that this book or 85321 can prepare you for every situation and problem you will come across. Additionally, for most of us mere mortals it is simply not possible for us to know everything that is required. Instead as a Systems Administrator must you know the important facts and be able to quickly discover any new information that you dont yet know. You must be able to diagnose and solve the problem even though you have never seen it before. This chapter examines the sources of information that a Systems Administrator might find useful including professional associations,books, magazines, and the Internet. It also provides some guidelines about how you might go about solving problems you have never seen before. While some of this information is Linux specific most of it is applicable to any operating system. As the semester progresses you should become familiar with and use most the information sources and the problem solving process presented here.

Other Resources
This chapter mentions a large number of Internet resources and includes the URLs. You can find an up to date listing of these links and other related links on the Links database on the 85321 Website (http://infocom.cqu.edu.au/85321/). Other resources which are related to this chapter include Online lecture 2 on the 85321 website HOW-TOs Online Troubleshooting Resources HOW-TO, Reading List HOW-TO and the Staying Updated mini-HOW-TO

Information Sources
The following sections first examine the range of information sources you have available as a Systems Administrator. Each of the following section deals with a different type of resource. As a trainee Sys Admin you should find yourelf starting to use these resources during your study. Learning how to answer your own questions is perhaps the most important thing you can take from this text. The information sources discussed in the following includes

David Jones (20.01.00)

Page 55

85321, Systems Administration

Chapter 2: Information Sources for Sys Admin

Professional associations There are quite a diverse list of professional assocations available which may be useful to a Systems Administrator. Books and magazines There are now a wide range of relevant books and magazines to chose from. Internet sources The Internet is perhaps the most comprehensive and up to date resources for Systems Administrators, if used correctly.

Professional organisations
Belonging to a professional organisation can offer a number of benefits including recognition of your abilities, opportunities to talk with other people in jobs similar to yours and a variety of other benefits. Most professional organisations distribute newsletters, hold conferences and many today have mailing lists and Web sites. All of these can help you perform your job. Professional organisations a Systems Administrator might find interesting include Systems Administrators Guild of Australia (SAGE-AU, http://www.sageau.org.au/), Systems Administrators Guild(SAGE) (the American version of SAGEAU, http://www.usenix.org/sage/), Australian UNIX Users Group (AUUG, http://www.auug.org.au/), Australian Computer Society (ACS, http://www.acs.org.au/), Usenix (http://www.usenix.org.au/), Internet Society of Australia (http://www.isoc-au.org.au/)

This list has a distinct Australian, UNIX, Internet flavour with just a touch of the USA thrown in. If anyone from overseas or from other factions in the computer industry (i.e. Novell, Microsoft) has a professional organisation that should be added to this list please let me know ([email protected]). The UNIX Guru Universe (UGU http://www.ugu.com/) is a Web site which provides a huge range of pointers to UNIX related material. It will be used throughout this chapter and in some of the other chapters in the text. Professional Associations The Resource Materials section on the 85321 Web site for week 1 has a page which contains links to professional associations and user organisations.

The SAGE groups

SAGE stands for Systems Administrators Guild and is the name taken on by a number of professional societies for Systems Administrators that developed
David Jones (20.01.00) Page 56

85321, Systems Administration

Chapter 2: Information Sources for Sys Admin

during the early 90s. There are national SAGE groups in the United States, Australia and the United Kingdom.

SAGE-AU
The Australian SAGE group was started in 1993. SAGE-AU holds an annual conference and distributes a bi-monthly newsletter. SAGE-AU is not restricted to UNIX Systems Administrators. Both SAGE and SAGE-AU have a presence on the WWW. The Professional Associations page on the 85321 Web site contains pointers to both.

UNIX User groups

There are various UNIX user groups spread throughout the world. AUUG is the Australian UNIX Users Group and provides information of all types on both UNIX and Open Systems. Usenix was one of the first UNIX user groups anywhere and is based in the United States. The American SAGE group grew out of the Usenix Association. Both Usenix (http://www.usenix.org/)and AUUG (http://www.auug.org.au/)have WWW sites. Both sites have copies of material from the associations newsletters. It should be noted that both user groups have gone beyond their original UNIX emphasis. This is especially true for Usenix which runs two important symposiums/conferences on Windows NT.

The ACS, ACM and IEEE

The ACS is the main professional computing society in Australia servicing people from all computing disciplines. The flavour of the ACS is much more business oriented than SAGE-AU. The ACS is also moving towards some form of certification of computing professionals and some jobs may require ACS membership. For more information refer to the WWW page (http://www.acs.org.au/). The Association for Computing Machinery (the ACM, http://www.acm.org/) is one of the largest American professional computing societies. Its publications are considerably more technical and wide ranging than the ACS. The Institute for Electrical and Electronics Engineers (IEEE, http://www.ieee.org/) also has a Computing Society (http://www.computer.org/).

Books and magazines

When a new computing person asks a technical question a common response will be RTFM. RTFM stands for Read The Fine (and other words starting with f) Manual and implies that the person asking the question should go away and look at documentation for the answer. Around five years ago RTFM for a Systems Administrator meant reading the on-line man pages, some badly written manual from the vendor or maybe, if
David Jones (20.01.00) Page 57

85321, Systems Administration

Chapter 2: Information Sources for Sys Admin

lucky, a Usenet newsgroup or two. Trying to find a book that explained how to use cron or how to set up NFS was a difficult task. Since then there has been an explosion in the number of books and magazines that cover Systems Administration and related fields. This is especially noticable over the last couple of years with the huge hype surrounding Linux and Open Source software. The following pages contain pointers to a number of different bibliographies that list books that may be useful. A Linux specific "magazine" which anyone with access to the 85321 CDROM/Website (or to the Linux Documentation Project) can read is the Linux Gazzette.

Bibliographies
UNIX, Systems Administration and related books. The Resource Materials section for week 1, on the 85321 Web site and CDROM, has a collection of pointers to books useful for 85321 and Systems Administrators in general.

OReilly books
Over the last few years there has been an increase in the number of publishers producing UNIX, Systems Administration and network related texts. However one publisher has been in this game for quite some time and has earned a deserved reputation for producing quality books. A standard component of the personal library for many Systems Administrators is a collection of OReilly books. For more information have a look at the OReilly Web site (http://www.ora.com/).

Magazines
There are now a wide range of magazines dealing with all sorts of Systems Administration related issues, including many covering Windows NT. Magazines The 85321 Web site contains pointers to related magazines under the Resource Materials section for week 1.

Internet resources
The Internet is by far the largest repository of information for computing people today. This is especially true when it comes to UNIX and Linux related material. UNIX was an essential part of the development of the Internet, while Linux could not have been developed without the ease of communication made possible by the Internet. If you have a question, a problem, need an update for some software, want a complete operating system or just want to have a laugh

David Jones (20.01.00)

Page 58

85321, Systems Administration

Chapter 2: Information Sources for Sys Admin

the Internet should be one of the first places you look as a Systems Administrator. So what is out there that could be of use to you? You can find software discussion forums, and information.

Each of these is introduced in more detail in the following sections.

The 85321 Website

The 85321 website (http://infocom.cqu.edu.au/85321/) contains mirrors, pointers and other resources related to some of the Internet related material discussed below. These resources will include a link database Chances are the links listed here will become out of date. The 85321 Website maintains a database of Linux and Systems Administration related links which include ratings. You can add links as to this database. a FAQ database The LDP and most newsgroups maintain lists of common questions (with answers) this database allows you to quickly search and view these questions. a Deja search interface The 85321 home page contains a form where you can perform a search of DejaNews.

How to use the Internet

By this stage it is assumed that you should be a fairly competent user of the Internet, the World-Wide Web, email, Usenet news and other net based resources. If you are a little rusty or haven't been introduced to many of these tools there are a large number of tutorials on the Internet that provide a good introduction. A good list of these tutorials is held on the Yahoo site (http://www.yahoo.com/).

Software on the Internet

There is a large amount of "free" UNIX software available on the Internet. It should be remembered that no software is free. You may not pay anything to get the software but you still have to take the time to install it, learn how to use it and maintain it. Time is money. GNU software (GNU is an acronym that stands for GNU's Not UNIX) is probably the best known "public-domain" software on the Internet. Much of the software, for example ls cd and the other basic commands, that comes with Linux is GNU software. In fact there is a trend amongst some people to call Linux, GNU/Linux, to reflect the amount of GNU software a Linux distribution uses.

David Jones (20.01.00)

Page 59

85321, Systems Administration

Chapter 2: Information Sources for Sys Admin

The Gnu Manifesto A copy of the GNU manifesto is available on the 85321 Web site and CDROM under the Resource Materials section for this week. The GNU website (http://www.gnu.org/) contains a lot more information about GNUs projects.

A good place to go to get the latest Open Source software (if you are based in Australia) is the AARNet mirror (http://www.mirror.aarnet.edu.au or ftp://mirror.aarnet.edu.au). It contains mirrors of a lot of popular Open Source software including GIMP, MySQL, Perl, Linux, Apache, KDE etc.

Discussion forums
Probably the biggest advantage the Internet provides is the ability for you to communicate with other people who are doing the same task. Systems Administration is often a lonely task where you are one of the few people, or the only one, doing the task. The ability to share the experience and knowledge of other people is a big benefit. Major discussion forums on the net include Usenet news

Mailing lists other discussion tools

Usenet news Once one of the most popular features of the Internet Usenet news has lost some of its popularity as the number of people on the Internet increases. Usenet news is a collection of discussion forums which are distributed and read with specialised software (i.e. You read Usenet news with a news reader). There are discussion forums on a wide range of topics from purely social through to very technical. A Systems Administrators technical interested is usually attracted to the large number of Linux/Systems Administration/Network related newsgroups. Newsgroups are a good place to ask questions, listen to other people and learn about topics. Some of the more useful newsgroups for this 85321 include
comp.os.linux.*

There are a large number of newsgroups under this heading discussing most Linux related topics, e.g. comp.os.linux.setup is used for discussion about installing and setting up a Linux box.
comp.unix.*

Another large collection of newsgroups talking about UNIX in particular.

David Jones (20.01.00)

Page 60

85321, Systems Administration

Chapter 2: Information Sources for Sys Admin

Useful groups include comp.unix.questions for general UNIX questions and comp.unix.admin for Systems Administration type questions.

aus.computer.linux

An Australian Linux newsgroup. http://www.linuxresources.com/online.html maintains a more detailed description and list of Linux newsgroups.

Just the FAQs

As you might imagine there are some questions which are asked again and again and again on newsgroups. Rather than repeat the answers to these questions again and again most newsgroups maintain a list of Frequently Asked Questions (FAQs). It is considered good practice when joining a newsgroup for the first time to read the the FAQs. It is a compulsory task before you ask a question on a newsgroup. You can access the FAQs for most newsgroups at http://www.faqs.org/. This site contains over 3300 seperate FAQs written by over 1250 authors covering 1700 newsgroups. The 85321 website also maintains a collection of many of the FAQs relevant to Systems Administration. Exercises
2.1.

There is a newsgroup called comp.os.unix. Like many newsgroups this group maintains an FAQ. Obtain the comp.unix.questions FAQ and answer the following questions - find out what the rc stands for when used in filenames such as .cshrc
/etc/rc.d/rc.inet1

- find out about the origins of the GCOS field in the /etc/passwd file

Google and Deja News

Google (http://www.google.com/) and DejaNews (http://www.deja.com/) are among the two most useful websites when it comes to finding technical information. Learn to use and enjoy them. Google is a Web search engine which uses technology to provide better ranking of Websites than other search engines. DejaNews is an archive and search engine of posts to Usenet News. The quickest way to find solutions to a lot of problems you will come across is to perform an appropriate search on DejaNews. Learn to use these sites.

Mailing lists
For many people the quality of Usenet News has been declining as more and more people start using it. One of the common complaints is the high level of beginners and the high level of noise. Many experienced people are moving towards mailing lists as their primary source of information since they often

David Jones (20.01.00)

Page 61

85321, Systems Administration

Chapter 2: Information Sources for Sys Admin

are more focused and have a better collection of subscribers and contributors. Mailing lists are also used by a number of different folk to distribute information. For example, vendors such as Sun and Hewlett Packard maintain mailing lists specific to their operating systems (Solaris and HP-UX). Professional associations such as SAGE-AU and SAGE also maintain mailing lists for specific purposes. In fact, many people believe the SAGE-AU mailing list to be the one of the best reasons for joining SAGE-AU as requests for assistance on this list are often answered within a few hours (or less). Mailing lists One good guide to all the mailing lists that are available is Liszt, mailing list directory (http://www.liszt.com/). The UNIX Gurus Universe also maintains a directory of mailing lists related to Sys Admin.

Other Discussion Forums

There are also other forums that may be useful for Systems Administrators and make use of technology other than Usenet news or mailing lists. These forums often use IRC or Web-based chat and bulletin board facilities. Over the last year or so Web-based bulletin board like systems have come to the fore. Examples include Slashdot, http://www.slashdot.org/ A bunch of people contribute links and information about what is happening in the nerd world. Includes a lot of interesting Linux related material. Linux Today, http://www.linuxtoday.com/ A slightly more business oriented version of Slashdot. A bit more serious but still a great information source. Freshmeat, http://www.freshmeat.net/ A place where people announce the latest releases of Open Source software

Internet based Linux resources

Linux would not have been possible without the Internet. The net provided the communications medium by which programmers from around the world could collaborate and work together to produce Linux. Consequently there is a huge collection of Internet based resources for Linux. The Linux Documentation Project The best place to start is the Linux Documentation Project (LDP). The aim of this project is to produce quality documentation to support the Linux community. The original LDP page is located at http://www.linuxdoc.org/ A mirror of the LDP pages is maintained on the 85321 Web site and a copy of these pages can be found on the 85321 CD-ROM.
David Jones (20.01.00) Page 62

85321, Systems Administration

Chapter 2: Information Sources for Sys Admin

A major source of information which the LDP provides are the HOW-TOs. HOW-TOs are documents which explain how to perform specific tasks as diverse as how to install and use StarOffice (a commercial office suite that is available free, and may well be on the 85321 CD-ROM) through to detailed information about how the Linux boot-prompt works. The HOW-TOs should be the first place you look for specific Linux information. Copies are available from the LDP Web pages. RedHat This version of the text is written as a companion for RedHat Linux. As a result it will be a common requirement for you find out information specific to RedHat Linux. The best source on the Internet for this information is the RedHat site, http://www.redhat.com/. Possibly the most important information source on the RedHat site are the updates/errata pages for the distribution of RedHat Linux you have installed. There will be errors and new software for RedHat Linux and RedHat are the best source. If you installed RedHat off a CD you may also find some of the updates and errata on it. Additionally RedHat provide three manuals with RedHat Linux 6.1. All three of the manuals are available as postscript files on the RedHat Linux CD. The three manuals are Red Hat Installation Guide A copy of this will be included in the package received by distance education students of CQU. This guide is designed to help you install RedHat Linux 6.1 onto your computer. Red Hat Linux Getting Started Guide Provides information you will likely need to configure your Linux box including an overview of Gnome. Gnome is the default desktop environment with RedHat Linux 6.1. RedHat Linux Reference Guide Contains references to a wide range of information you may need while using your computer.

Problem Solving
I will guarantee that you will have problems with Linux while you are attempting the tasks and reading about the concepts introduced in this text. Most of the time these problems will be unlike anything you have ever seen before. You probably wont have any idea what the actually problem is. You may have difficulty even starting to describe it. Dont PANIC!!!! Every Systems Administrator, every computer user has faced this same problem. The difference between a bad Systems Administrator and a good one is the ability to problem solve. If you can learn to solve problems you have never faced before then you can do anything.

David Jones (20.01.00)

Page 63

85321, Systems Administration

Chapter 2: Information Sources for Sys Admin

Guidelines for solving problems

The "Linux Installation and Getting Started Guide" (part of the Linux Documentation Project) provides the following guidelines for solving problems. They are a good start. Remain calm. Never ever attempt to solve anything computer related while you are upset, angry, sad, tired or emotional in any way. Any of these feelings will adversely affect your ability to solve the problem. In other words they are likely to make things worse. Learn to appreciate self-reliance The key to problem solving is that you have to do it. If you are continually relying on other people to solve your problems you will never learn anything. Try and solve your problems first. If you cant solve your problems talk to someone about what you are doing and why you are doing it. Try and figure out why you cant solve problems. The idea is that you have to do this before you can start problem solving on your own. Consult all available sources of information Most people, especially in the Linux world, are more than happy to help out with a problem. However, they can get very upset if the question you are asking has already been answered in some documentation or on the Web. Know the best places to look There are a huge number of different places from which you can get information about Linux and Systems Administration. The same applies to Windows NT. If you know which sites are best for which sort of information you can significantly cut down your search time. Refrain from asking spurious questions Linux has the best support mechanism of any operating system and the best thing about it is that it is free. However, there are accepted codes of behaviour and one of them is "Dont ask stupid questions". When asking for help be polite, terse and informative "I cant boot my Linux computer" is not informative. If you dont provide sufficient information no-one will be able to offer help. At the other extreme, if you provide too much information people wont be bothered to read it. Lastly, no-one likes a rude person. If you are rude people are likely to be rude back again and not help. Contribute to the community The number of people contributing to Linux newsgroups, the LDP and other areas is the main reason why Linux support is so good. If you know the answer to a problem share it with other people and help make the support even better. These guidelines are a good start. The following are some additional ones Ive developed over the last few years. Understand what the problem is. This may seem to be a bit obvious. However, 5 years of teaching 85321 has shown that the most common reason students cant solve their problems
Page 64

David Jones (20.01.00)

85321, Systems Administration

Chapter 2: Information Sources for Sys Admin

is that they dont understand what the problem is. If you dont know what the problem is find out or following the next suggestion. Break the problem down into smaller steps If the problem is too complex, too large or you dont understand it try breaking it down into smaller steps so you can solve (or understand) those. Dont get to close, walk away There will be times when the best thing you can do is walk away and do something else for a while. Any person with experience in the computing field should be able to tell a story about how a solution to a frustrating problem popped into their head while mowing the lawn, watching television or having a shower. Some distance from the problem can provide the bit of perspective you needed to figure out the problem. Talk to people. A group of people will always be able to solve more problems than a single person. More people means more experience and different perspectives. If you dont know the solution chances are someone else does. Record the solution This could be considered as contributing to the community but it also serves a much more selfish reason. If you have faced the problem once then chances are it you, or another Systems Administrator, will face it again. If you have documented the solution to the problem then solving it the 2nd, 3rd or 4th time will be much quicker.

Examples of solving problems

The 2nd 85321 online lecture includes three examples of how to use the resources and approaches introduced above to solve problems.

Conclusions
A lot of time spent by Systems Administrators is consumed attempting to solve new problems. A Systems Administrator must be able to solve new problems. Being aware of the available information sources and their relative merits is an important step towards being able to solve these problems. While the information presented in this chapter may be useful it is experience which is the most useful tool in solving problems. Take the time to gain this experience.

Review Questions
2.1 Use the information sources here to solve some of the problems you are currently having. If you arent having problems, tind a question from one of the Linux or UNIX newsgroups. 2.2. Examine the errata list for your version of RedHat Linux. Do any of these errata appear important to your system?

David Jones (20.01.00)

Page 65

85321, Systems Administration

Chapter 3: Using UNIX

Chapter
Using UNIX

Introduction
A Systems Administrator not only has to look after the computers and the operating system, they also have to be the expert user (or at least a very knowledgeable user) of their systems. When other users have problems where do they go? The documentation? Online help facilities? No, they usually go to the Systems Administrator. Adding to the importance of the material covered in the next few chapters is that a number of the topics introduced here are the foundations on which some of the more complex Systems Administration topics are built. If you dont understand these concepts now you will have problems later on. The following reading aims to start you on the road to becoming an expert UNIX user. Becoming a UNIX guru can only be achieved through a great deal of experience so it is important that you spend time using the commands introduced in this chapter.

Other Resources
Resources explaining the basics about using Linux and UNIX are quite numerous. Some of the other resources which mention similar concepts to this chapter include The RedHat Manuals RedHat 6.1 comes with three manuals. This chapter refers to some of these manuals as a source for more information. Online lecture 4 Included on the 85321 website/CD-ROM this lecture with slides and audio covers complementary material to this chapter. Linux Installation and Getting Started Guide One of the guides included with the Linux Documentation Project includes some basic information. A copy of the LDP is available on the 85321 website/CD-ROM.

What you need to learn

All of the concepts and material you will use to learn how to use Linux is not included in this book. Throughout this chapter you will be referred a collection of documents and web pages. It is important that you actually read this material and more importantly you should attempt to practice the commands and practices you learn about. If you dont use it you lose it.

David Jones (20.01.00)

Page 66

85321, Systems Administration

Chapter 3: Using UNIX

In order to be able to really understand the material introduced later in this chapter and to be able to perform the required tasks with a minimum of effort you MUST become familiar with the following How to get around the Linux file hierarchy. Using the GUI "explorer-like" tools provided by Gnome and KDE are not sufficient. You must be familiar with using commands like cd ls mkdir rm ls cp How to get the most out of the user interface you are provided with. Making use of the GUI, while not a replacement for the command line, will make some task easier. Be able to use simple command line tools Im repeating it, just in case you ignored it the first time. You MUST BE ABLE to use the simple UNIX commands such as ls cd, mkdir rm ls cp etc Be able to use the vi editor. As with the UNIX command line many of you will question why you need to use vi. The simple reason is that it will make your life much easier later in the semester if you learn how to use vi now. Understand the Linux file permissions system. This is especially essential. An understanding of the Linux file permissions system is an absolute necessity for when you move onto the more complex Systems Administration concepts introduced in later chapters. If you dont understand this now you will have major problems later on. Be able to manipulate and view the processes currently running. As with file permissions the ability to manipulate and view the processes on a Linux box is an essential skill for a Systems Administrator.

Introductory UNIX
Five years ago this section used to be quite easy for the majority of people reading this text. However, since then the explosion in Windows and other GUIs means that most people have little or no experience with using the command line. Some of you may not even now what the command line is!!! The command line is a name given to the text-based interface which was common a few years ago and is still present in the form of the MS-DOS prompt in the Windows world and command-line shells in the UNIX world. This interface uses a process something like this computer displays a prompt use types a command, usually in the format command_name a list of parameters computer tries to carry out that command and displays any output computer displays a prompt again (and we loop back to the top)

Im sorry to say but as a Systems Administrator you have to know how to use the command line. This means you have to forget about that nice GUI
David Jones (20.01.00) Page 67

85321, Systems Administration

Chapter 3: Using UNIX

provided by Explorer and start to understand the structure of files and directories used by Linux.

Why do I need to know the command line?

Some possible answers to this question include The GUI isnt always available. Unlike Windows NT a GUI is not a compulsory part of UNIX. This is one of the reasons why a small 386 running Linux can act as the server for a small organisation. It also means that there will be times as a Systems Administrator that you will have to perform tasks without a GUI. Those times are generally when something has broken. You wont have the time to learn how to use the command-line then, take the time to do it now. The command line is often more efficient and powerful. There are a wide number of tasks which you will have to perform in your computing career that are not suited to using a GUI. These tasks can be done quicker and easier using the command line.

The last reason for those of you studying 85321 is that your ability to use the command line is assessable. If you dont know how to use it you will lose marks.

How do I learn all this stuff?

The simple answer is practice. You cant learn this material without experience. First you need to read and listen to the material pointed to below. Then you must take the time to perform the tasks set and also possible a number of others until you are comfortable with using the command line. Taking the time to do this now will save you time later. Basic UNIX There is a wide range of material on the Internet which will introduce you to the basics of using UNIX. The following is a list of some of those available from the 85321 Website/CD-ROM. Please use the resources which best suit you
7.1.

Online Lecture 2 Produced for the 1999 offering of 85321 this lecture covers using the command line, vi and the file hierarchy The Linux Installation and Getting Started Guide This guide was produced as part of the Linux Documentation Project (a mirror of this project is included on the 85321 Website/CD-ROM). It has a Linux tutorial which covers much of the basic material

7.2.

David Jones (20.01.00)

Page 68

85321, Systems Administration

Chapter 3: Using UNIX

7.3.

The Red Hat Linux Getting Started Guide Produced by Red Hat and included on the 85321 Website/CD-ROM in PDF and HTML formats this guide also covers much of the introductory material you will need.

Exercises
3.1.

What UNIX commands would you use to - change to your home directory - display the list of files in the current directory - display my name is fred onto the screen - copy the file tmp.dat from the current directory to the directory data underneath your home directory and after the file has been copied delete it What will the following UNIX commands do? Dont execute a UNIX command if you arent sure what it is going to do. In particular do not try to execute the first command below.
rmdir ~ cat /etc/passwd ls ../../fred/doc/tmp

3.2.

3.3.

Indicate which of the following paths are full or relative a./root/ b.../root/ c./usr/../root

d./home/david/

3.4.

Assuming you are currently in the /home/david/tmp/ directory write the full path of your final location if you perform the following commands a.cd ../85321/ b.cd /usr/lib c.cd ~/85321 Answer the following questions a.Where would you find the home directory for the root user? b.Where would you store some temporary files? c.Where do you normally find the home directories of "normal" users? d.Assuming you were currently in the directory containing the boot configuration files, how would you change into the directory containing the system configuration files and scripts?

3.5.

UNIX Commands are programs

The UNIX commands that have been introduced so far are stored on a UNIX computer as executable files. All the commands on a UNIX systems are either stored on the hard-drive as executable files or are understood by a shell (more on these in a later chapter). Most of the standard commands will be stored in standard binary directories such as /bin /usr/bin /usr/local/bin. On my system running RedHat version 6.1 there are 1494 different files in the directories /bin, /usr/bin and /usr/sbin. Which means over 1494 different commands (or there abouts).

David Jones (20.01.00)

Page 69

85321, Systems Administration

Chapter 3: Using UNIX

vi
A major task of any user of a computer is editing text files. For a Systems Administrator of a UNIX system manipulation of text files is a common task due to many of the system configuration files being text files. The most common, screen-based UNIX editor is vi. The mention of vi sends shudders through the spines of some people, while other people love it with a passion.
vi is difficult to learn, however it is also an extremely powerful editor which can save a Systems Administrator a great deal of time.

As you progress through this subject you will need an editor. vi is an anachronistic antique of an editor hated by most people. So why should you use it? Reasons include it is very powerful, How many editors do you know that can take the first 20 characters of every line in a file and swap them with the second set of 20 characters (something Ive had to do) it is the only screen editor available on every UNIX system There will be times when a Systems Administrator cannot use a full screen editor. At times like this you must resort to single line editors like ed and ex. vi grew out of the ex editor and so uses many of the same commands. Learning and using these commands in vi can save you time later on.

As a result of all this it is strongly suggested that you use vi wherever possible in studying for this unit. Early on you will find using vi a hassle but sticking with it will be worthwhile in the end.

An introduction to vi
Most people when confronted with vi for the first time are put off by its completely foreign nature and lack of any queues about what to do. vi actually uses a very simple "model of operation". Central to this is the fact that vi is a modal editor. This means vi has a number of different modes and the same action can have completely different meaning in different modes. vi modes vi can be said to have three modes command mode This is the default mode vi is in when you start it up. In this mode most of the keys on the keyboard perform vi commands. For example, hitting the e key during vi command mode moves the cursor onto the next word. Use the list of vi commands in any of the vi command references discussed below to find out more. insert mode This is the mode in which vi behaves most like other editors. If you hit the k key it will insert k into the current location of the cursor and move the cursor on.

David Jones (20.01.00)

Page 70

85321, Systems Administration

Chapter 3: Using UNIX

ex mode In ex (sometimes called colon mode) you get to access a range of commands from the ex editor (and you thought vi was hard to use). A common one you will use is :wq which writes/saves the current file and then quits vi (wq).

vi Transitions Knowing about the vi modes is no good unless you know how to go from one mode to another. For example, you cant actually type anything into a text file you are creating without knowing how to go from command mode to insert mode. Common transitions include command to insert A number of vi commands take you from command to insert modes (e.g. i o O) insert to command Youll do this transition when you want to save a file (usually). Hitting the ESC key is usually enough to achieve this command to ex Simply hitting the : (colon) key will put you into ex mode. You will know this because the colon will appear at the bottom of the screen. ex to command Simply hitting enter at the end of an ex command takes you back into command mode. Using vi The section for week 2 in the Link database on the 85321 Web page contains a number of resources which introduce you to vi. This includes the 4th online lecture which has a number of slides and examples of using vi. vi, vim and ^Ms A common problem in the last couple of years for 85321 students has shown itself up as shell scripts which cant run (youll be getting to shell scripts in a couple of weeks). The problem usually occurs when the student copies a text file created under Windows to Linux. The cause of the problem is that UNIX and Microsoft indicate the end of the line in different ways carriage return, line-feed Used by Microsoft operating systems line feed Use by Linux.

The extra character, the carriage return, causes problems in some situations, e.g. When you want to run the text file as a shell script.

David Jones (20.01.00)

Page 71

85321, Systems Administration

Chapter 3: Using UNIX

The solution is to remove the extra characters. One method is to use vi. The trouble is that by default vim, the version of vi on Linux, is smart enough to hide the carriage returns. If you have a text file, which has carriage returns in it, and you want to see the carriage returns you have to do the following
vi -b filename

The -b causes vi to work in binary mode and you will now be able to see the carriage returns which look like ^M and appear at the end of the line. Carriage return is actually one character. ^M is the standard UNIX way of representing a single control character. If you try to delete the ^M with the x command you will find that there is only on character. To delete all the carriage returns in a file you can use the following command
1,$s/^M//g

Where you dont type the ^ character and then the M character. Instead you hold the CONTROL key down and hit the c key and then hit the m key. What this command does should become clear when we talk about regular expressions in a later chapter.

UNIX commands
A UNIX system comes with hundreds of executable commands and programs . Typically each of these programs carries out a particular job and will usually have some obscure and obtuse name that means nothing to the uninitiated. That said the names of most of these commands actually do make some sort of sense once you have a bit of knowledge. In the following you are introduced to the philosophy and format of UNIX commands. It is also emphasised that there is almost always going to be a UNIX command (or a combination of them) to perform the task you wish to accomplish. You need to become familiar with how to find out about the available commands.

Philosophy of UNIX commands

There are no set rules about UNIX commands however there is a UNIX philosophy that is used by many of the commands. small is beautiful, UNIX provides the mechanisms to join commands together so commands should do one thing well. 10 percent of the work solves 90 percent of the problems, UNIX was never designed to solve all problems, it was designed to solve most requirements without too much hassle on the programmers part. solve the problem, not the machine, Commands should ignore any machine specific information and be portable.

David Jones (20.01.00)

Page 72

85321, Systems Administration

Chapter 3: Using UNIX

solve at the right level, and you will only have to do it once. The key to UNIX problem solving is only to do it once e.g. pattern matching is only implemented once, in the shell, not in every command.

One of the central tenants of the UNIX command philosophy is to provide a flexible, adaptable toolbox approach to solving problems. The idea is not to provide a single large program which does everything. Instead you have small, purpose built commands which can be easily combined to perform much larger tasks. An evolution rather than creation approach to solving problems.

UNIX command format

UNIX commands all use the following format
command_name -switches parameter_list

Component
command_name

Explanation the name of the actual command, generally this is the name of the executable program that is the command The - symbol is used to indicate a switch. A switch modifies the operation of a command. the list of parameters (or arguments) that the command will operate on, could be 0, 1 or more parameters, parameters are separated by white space characters (space, TAB)
Table 3.1 UNIX command format

-switches

parameter_list

Please note: there must be spaces between each component of a UNIX command line. Under MS-DOS it was possible to perform commands like cd/dos Change the current directory into the /dos directory.

Under UNIX this command will be interpreted as run the command cd/dos. This means UNIX will normally try to find an executable file called cd/dos. The UNIX shell (the command which interprets the command line and tries to execute commands) uses the space character to tell the difference between the different components of the command line. Under UNIX the command would have to be cd /dos

Example commands
ls -l

The switch -l is used to modify the action of the ls command so that it displays a long listing of each file.
ls -l /etc/passwd / /var

Commands can take multiple parameters.

ls -ld /etc/passwd / /var

Multiple switches can also be used.

David Jones (20.01.00)

Page 73

85321, Systems Administration

Chapter 3: Using UNIX

Linux commands take multiple arguments Unlike MS-DOS, UNIX commands can take multiple arguments. However, the multiple parameters must be separated by space characters. Exercises
3.6.

One of your users has created a file called -tmp? (The command cat /etc/passwd > -tmp will do it.) They want to get rid of it but cant. Why might the user have difficulty removing this file? How would you remove the file? You might have to refer to the online help section below to find the answer.

A command for everything

A fairly intelligent and experienced would be computer professional has just started using UNIX seriously (he was a student in the very first offering of this subject). He gets to a stage where he wants to change the name of some files. Being an MS-DOS junkie from way back what command does he look for? The rename command of course. It doesnt work! "Thats a bit silly!", he thinks, "You would think that UNIX would have a rename command." It just so happens that this person has just completed a C programming subject in which one of the assignments was to write a rename command. So he spends the next day trying to write and compile this program. After much toil and trouble he succeeds and follows good administration policy and informs all the other students of this brand new wonderful program he has written. He goes into great detail on how to use the command and all the nice features it includes. They all write back to tell him about the UNIX command mv (the move command) that is the UNIX command that is equivalent to rename. The moral of the story The moral of this story is that if you want to do something under UNIX, then chances are that there is already a command to do it. All you have to do is work out what it is.

Online help
UNIX comes with online help called man pages. Man pages are short references for commands and files on a UNIX system. They are not designed as a means by which newcomers to UNIX can learn the commands. Instead they are a reference to the command to be used by someone who understands the basics of UNIX and UNIX commands. The man pages are divided into different sections. Table 3.2 shows the sections that Linux uses. Different versions of Linux use slightly different sections.

David Jones (20.01.00)

Page 74

85321, Systems Administration

Chapter 3: Using UNIX

Section number 1 2 3 3c 3s 3m 3f 3x 4 5 6 7 8 user commands system calls Library functions

Contents

standard C library standard I/O library arithmetic library Fortran library special libraries special files file formats games miscellaneous administration and privileged commands
Table 3.2 Manual Page Sections

Using the manual pages

To examine the manual page for a particular command or file you use the man command. For example if you wanted to examine the man page for the man command you would execute the command man man.

Is there a man page for...

The command man -k keyword will search for all the manual pages that contain keyword in its synopsis. The commands whatis and apropos perform similar tasks. Rather than search through all the manual pages Linux maintains a keyword database in the file /usr/man/whatis. If at any stage you add new manual pages you should rebuild this database using the makewhatis command. The -K switch for the man command forces it to search through all of the manual page for the work. You should realise that with the size and number of manual pages this operation can take quite a while. If there is a file you wish to find out the purpose for you might want to try the f option of the man command.

man page format

groff

Each manual page is stored in its own file formatted (under Linux) using the command (which is the GNU version of nroff). The files can be located

David Jones (20.01.00)

Page 75

85321, Systems Administration

Chapter 3: Using UNIX

in a number of different directories with the main manual pages located under the /usr/man directory. Under /usr/man you will find directories with names mann and catn. The n is a number that indicates the section of the manual. The files in the man directories contain the groff input for each manual page. The files in the cat directories contain the output of the groff command for each manual page. Generally when you use the man command the groff input is formatted and displayed on the screen. If space permits the output will be written into the appropriate cat directory.
3.7.

What commands would you use to do the following 1.View mkdir(2) 2.Find the command to print a file 3.How many different manual pages exist for mkdir 4.Describe the contents of section 8 of the manual pages.

HTML versions of Manual Pages

The 85321 website contains a manual page section which contains a collection of manual pages in HTML format.

Some UNIX commands

There are simply too many UNIX commands for this chapter to introduce all, or even most of them. The aim of the following is to show you some of the basic commands that are available. To find the remainder you will have to discover them for yourself. One method for becoming more familiar with the available commands is to look at the filenames in the /bin /usr/bin /usr/local/bin directories, These are the binary directories which contain the executable programs which are the UNIX commands. take the filename and look at the manual page Each of the commands will have a manual page which will explain what the command does and how you can use it.

The commands introduced in this table can be divided into categories based on their purpose identification commands, These commands identify the user or the system. simple commands, Perform some simple, non-specific task. filters. Filters take some input, modify it and then output it.

David Jones (20.01.00)

Page 76

85321, Systems Administration

Chapter 3: Using UNIX

Command
date

Purpose Display the current time and date Display a large banner Displays your current username Display the contents of a file a page at a time Display the last few lines of a file Remove duplicate lines from a file join columns of files together Display all lines in a file containing a patter
Basic

Command
who

Purpose display who is currently on the computer display a calendar display the contents of a file display the first few lines of a file sort the content of a file into order remove columns of characters from a file translate specific characters count the number of characters, words and lines in a file

banner

cal

whoami

cat

more

and less

head

tail

sort

uniq

cut

paste

tr

grep

wc

Table 3.3 UNIX commands

Identification Commands who

Displays a list of people currently logged onto the computer.
dinbig:/$ who david tty1 Feb 5 14:27

whoami
Displays who the computer thinks you are currently logged in as.
dinbig:/$ whoami david

David Jones (20.01.00)

Page 77

85321, Systems Administration

Chapter 3: Using UNIX

uname Displays information about the operating system and the computer on which it is running
[david@beldin david]$ uname Linux [david@beldin david]$ uname a Linux beldin.cqu.edu.au 2.0.31 #1 Sun Nov 9 21:45:23 EST 1997 i586 unknown

Simple commands
The following commands are simple commands that perform one particular job that might be of use to you at some stage. There are many others youll make use of. Only simple examples of the commands will be shown below. Many of these commands have extra switches that allow them to perform other tasks. You will have to refer to the manual pages for the commands to find out this information.

date
Displays the current date and time according to the computer.
dinbig:/$ date Thu Feb 8 16:57:05 EST 1996

banner
Creates a banner with supplied text.
dinbig:/$ banner -w30 a ## ###### ## ## ## ### # # # ## ## ## ########### ##

cal
Display a calendar for a specific month. (The Linux version might not work).
bash$ cal 1 1996 January 1996 S M Tu W Th F S 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

David Jones (20.01.00)

Page 78

85321, Systems Administration

Chapter 3: Using UNIX

Filters
Filters are UNIX commands that take input or the contents of a file, modify that content and then display the result on output. Later on in this chapter you will be shown how you can combine these filters together to manipulate text.

cat
The simplest filter. cat doesnt perform any modification on the information passed through it.
bash$ cat /etc/motd Linux 1.2.13.

more and less

These filters display their input one page at a time. At the end of each page they pause and wait for the user to hit a key. less is a more complex filter and supports a number of other features. Refer to the man page for the commands for more information.

head and tail

head

and tail allow you to view the first few lines or the last few lines of a

file. Examples
head chap1.html

Display the first 10 lines of chap1.html

tail chap1.html

display the last 10 lines of chap1.html

head -c 100 chap1.html

display the first 100 bytes of chap1.html

head -n 50 chap1.html

display the first 50 lines of chap1.html

tail -c 95 chap1.html

display the last 100 bytes of chap1.html sort sort The sort command is used to sort data using a number of different criteria outlined in the following table.

David Jones (20.01.00)

Page 79

85321, Systems Administration

Chapter 3: Using UNIX

Switch
-r -n

Result sort in descending order (default is ascending) sort as numbers (default is as ASCII characters) When sorting numbers as numbers 100 is greater than 5. When sorting them as characters 5 is greater than 100. eliminate duplicate lines skip number fields specify character as the field delimiter
Table 3.4 Switches for the sort command

-u +numbern -tcharacter

Examples The following examples all work with the /etc/passwd file. /etc/passwd is the file that stores information about all the users of a UNIX machine. It is a text file with each line divided into 7 fields. Each field is separated by a : character. Use the cat command to view the contents of the file.
sort /etc/passwd

sort in order based on the whole line

sort -r /etc/passwd

reverse the order of the sort

sort +2n -t: /etc/passwd

sort on third field, where field delimiter is : (skip the first two fields)
sort +2n -t: -n /etc/passwd

same sort but treat the field as numbers not ASCII characters
uniq

uniq is used to find or remove and duplicate lines from a file and display what is left onto the screen. A duplicate to uniq is where consecutive lines match exactly. sort is often used to get the duplicate lines in a file into consecutive order before passing it to uniq. Passing a file from one command to another is achieved using I/O redirection which is explained in a later chapter. Examples
uniq names

remove duplicate lines from names and display them on the screen
uniq names uniq.names

remove duplicates lines from names and put them into uniq.names
uniq -d names

display all duplicate lines

David Jones (20.01.00) Page 80

85321, Systems Administration

Chapter 3: Using UNIX

tr

Used to translate specified characters into other characters. tr is used in conjunction with I/O redirection which is explained in the next chapter. In the examples below the < character is an I/O redirection character. Examples
tr a z < /etc/passwd translate all as to zs in /etc/passwd

and display on the screen character

tr [A-Z] [a-z] < /etc/passwd translate any character in between A-Z into the equivalent between a-z. (make all upper-case characters lower case) tr -d < /etc/passwd

delete any single space characters from the file

cut

Is used to "cut out" fields from a file. Try cut -c5-10 /etc/passwd. This will display all the characters between the 5th and 10th on every line of the file /etc/passwd. The following table explains some of the switches for cut Switch
-cRANGE -dcharacter -fRANGE

Purpose cut out the characters in RANGE specify that the field delimiter is character cut out the fields in RANGE
Table 3.5 Switches for the cut command

RANGE

used by the -f and -c switches can take the following forms

number-

get all from character or field number to the end of the line
number-number2

get all from character or field number to character or field number2

number,number2

get characters or fields number and number2 And combinations of the above. Examples
cut -c1 /etc/passwd

get the first character from every line

cut -c1,5,10-20 /etc/passwd

get the first, fifth character and every character between 10 and 20

David Jones (20.01.00)

Page 81

85321, Systems Administration

Chapter 3: Using UNIX

cut -d: -f2 /etc/passwd

get the second field

cut -d: -f3- /etc/passwd

get all fields from the third on

paste

This command performs the opposite task to cut. It puts lines back together. Assume we have two files
names george fred david janet addresses 55 Aim avenue 1005 Marks road 5 Thompson Street 43 Pedwell road

To put them back together wed use the command

bash$ paste names addresses george 55 Aim avenue fred 1005 Marks road david 5 Thompson Street janet 43 Pedwell road

The two fields have been separated by a tab character. To use a different character you use the -d switch.
bash$ paste -d: names addresses george:55 Aim avenue fred:1005 Marks road david:5 Thompson Street janet:43 Pedwell road

To paste together lines from the same file you use the -s switch.
bash$ paste -s names george fred david janet

grep grep stands for Global Regular Expression Pattern match. It is used to search a file for a particular pattern of characters.

grep david /etc/passwd

display any line from /etc/passwd that contains david To get the real power out of grep you need to be familiar with regular expressions which are discussed in more detail in a later chapter.

David Jones (20.01.00)

Page 82

85321, Systems Administration

Chapter 3: Using UNIX

wc

Used to count the number of characters, words and lines in a file. By default it displays all three. Using the switches -c -w -l will display the number of characters, words and lines respectively.
bash$ wc /etc/passwd 19 20 697 /etc/passwd bash$ wc -c /etc/passwd 697 /etc/passwd bash$ wc -w /etc/passwd 20 /etc/passwd bash$ wc -l /etc/passwd 19 /etc/passwd

For the following exercises create a file called phone.book that contains the following
george!2334234!55 Aim avenue fred!343423!1005 Marks road david!5838434!5 Thompson Street janet!33343!43 Pedwell road

The field delimiter for this file is ! and the fields are name, phone number, address. Exercises
3.8.

What command would you use to (assume you start from the original file for every question) 1. sort the file on the names 2. sort the file in descending order on phone number 3. display just the addresses 4. change all the ! characters to :
5. display the first line from the file 6. display the line containing davids information 7.What would effect would the following command paste -d: -s phone.book

have

3.9.

A University student database system must produce a number of files containing information about students in a class. One such example CSV file with 6 fields: student number, surname, firstname, grade (F,P,C,D,HD), mark and degree code is available from the 85321 website/CD-ROM at the URL http://infocom.cqu.edu.au/85321/Resources/Lectures/6/results.csv Using previous descriptions in the lecture, the 85321 text book and the Linux manual pages come up with commands to perform the following tasks on this example file 1.count the number of students in the class (hint: there is one student per line in the file) 2.display the data file one line at a time 3.get a list of all the student numbers in the class 4.Find all students who received HDs

David Jones (20.01.00)

Page 83

85321, Systems Administration

Chapter 3: Using UNIX

Getting more out of filters

The filters are a prime example of good UNIX commands. They do one job well and are designed to be chained together. To get the most out of filters you combine them together in long chains of commands. How this is achieved will be examined in a later chapter when the concept of I/O redirection is introduced. I/O redirection allows you to count the number of people on your computer who have usernames starting with d by using the grep command to find all the lines in the /etc/passwd file that start with d and pass the output of that command to the wc command to count the number of matching lines that grep found. How you do this will be explained next week.

Conclusions
In this chapter you have been provided a brief introduction to the philosophy and format of UNIX commands. In addition some simple commands have been introduced including account related commands
login passwd exit

file and directory manipulation commands

cd ls rm mv mkdir

some basic commands

date who banner cal

some filters
cat more less head tail sort uniq cut paste tr grep

A Systems Administrator has to be a "guru". An expert user of their system. A Systems Administrator should not only to be able to get the most out of the system but also to be able to explain and assist other users.

David Jones (20.01.00)

Page 84

85321, Systems Administration

Chapter 4: The File Hierarchy

Chapter
The File Hierarchy

Introduction
Why?
Like all good operating systems, UNIX allows you the privilege of storing information indefinitely (or at least until the next disk crash) in abstract data containers called files. The organisation, placement and usage of these files comes under the general umbrella of the file hierarchy. As a system administrator, you will need to be very familiar with the file hierarchy. You will use it on a day to day basis as you maintain the system, install software and manage user accounts. At a first glance, the file hierarchy structure of a typical Linux host (we will use Linux for the basis of our discussion) may appear to have been devised by a demented genius whod been remiss with their medication. Why, for example, does the root directory contain something like:
bin boot dev etc home lib lost+found mnt proc root sbin tmp usr var

Why was it done like this? Historically, the location of certain files and utilities has not always been standard (or fixed). This has lead to problems with development and upgrading between different "distributions" of Linux [Linux is distributed from many sources, two major sources are the Slackware and Red Hat package sets]. The Linux directory structure (or file hierarchy) was based on existing flavours of UNIX, but as it evolved, certain inconsistencies developed. These were often small things like the location (or placement) of certain configuration files, but it resulted in difficulties porting software from host to host. To combat this, a file standard was developed. This is an evolving process, to date resulting in a fairly static model for the Linux file hierarchy. In this chapter, we will examine how the Linux file hierarchy is structured, how each component relates to the overall OS and why certain files are placed in certain locations.

David Jones (20.01.00)

Page 85

85321, Systems Administration

Chapter 4: The File Hierarchy

Linux File System Standard The location and purposes of files and directories on a Linux machine are defined by the Linux File Hierarchy Standard. The offical website for the Linux File Hierarchy Standard is http://www.pathname.com/fhs/

The important sections

The root of the problem
The top level of the Linux file hierarchy is referred to as the root (or /). The root directory typically contains several other directories including: Directory
bin/ boot/ dev/ etc/ home/ lib/ Lost+found/ mnt/ proc/

Contains Required Boot-time binaries Boot configuration files for the OS loader and kernel image Device files System configuration files and scripts User/Sub branch directories Main OS shared libraries and kernel modules Storage directory for "recovered" files Temporary point to connect devices to Pseudo directory structure containing information about the kernel, currently running processes and resource allocation Linux (non-standard) home directory for the root user. Alternate location being the / directory itself System administration binaries and tools Location of temporary files r Difficult to define - it contains almost everything else including local binaries, libraries, applications and packages (including X Windows) Variable data, usually machine specific. Includes spool directories for mail and news
Table 4.1 Major Directories

root/

sbin/ tmp/ usr/

var/

Generally, the root should not contain any additional files - it is considered bad form to create other directories off the root, nor should any other files be placed there.

David Jones (20.01.00)

Page 86

85321, Systems Administration

Chapter 4: The File Hierarchy

Why root? The name root is based on the analogous relationship between the UNIX files system structure and a tree! Quite simply, the file hierarchy is an inverted tree. I can personally never visiualise an upside down tree what this phrase really means is that the top of the file heirarchy is at one point, like the root of a tree, the bottom is spread out, like the branches of a tree. This is probably a silly analogy because if you turn a tree upside down, you have lots of spreading roots, dirt and several thousand very unhappy worms! Every part of the file system eventually can be traced back to one central point, the root. The concept of a root structure has now been (partially) adopted by other operating systems such as Windows NT. However, unlike other operatings systems, UNIX doesn't have any concept of drives. While this will be explained in detail in a later chapter, it is important to be aware of the following: The file system may be spread over several physical devices; different parts of the file heirarchy may exist on totally separate partitions, hard disks, CDROMs, network file system shares, floppy disks and other devices. This separation is transparent to the file system heirarchy, user and applications. Different parts of the file system will be connected (or mounted) at startup; other parts will be dynamically attached as required. The remainder of this chapter examines some of the more important directory structures in the Linux file hierarchy.

Homes for users

Every user needs a home...
The /home directory structure contains the the home directories for most login-enabled users (some notable exceptions being the root user and (on some systems) the www/web user). While most small systems will contain user directories directly off the /home directory (for example, /home/jamiesob), on larger systems is common to subdivide the home structure based on classes (or groups) of users, for example:
/home/admin /home/finance /home/humanres /home/mgr /home/staff # # # # # Administrators Finance users Human Resource users Managers Other people

David Jones (20.01.00)

Page 87

85321, Systems Administration

Chapter 4: The File Hierarchy

Other homes?
/root /root

is the home directory for the root user. If, for some strange reason, the directory doesnt exist, then the root user will be logged in in the / directory - this is actually the traditional location for root users. There is some debate as to allowing the root user to have a special directory as their login point - this idea encourages the root user to set up their .profile, use "user" programs like elm, tin and netscape (programs which require a home directory in which to place certain configuration files) and generally use the root account as a beefed up user account. A system administrator should never use the root account for day to day user-type interaction; the root account should only be used for system administration purposes only. Be aware that you must be extremely careful when allowing a user to have a home directory in a location other than the /home branch. The problem occurs when you, as a system administrator, have to back-up the system - it is easy to miss a home directory if it isnt grouped with others in a common branch (like /home).

/usr

and /var

And the difference is...

It is often slightly confusing to see that /usr and /var both contain similar directories:
/usr X11R6 bin dict doc etc /var catman lib local lock

games i486-linux-libc5 include info lib log nis

libexec local man sbin share spool tmp

src tmp

preserve run

It becomes even more confusing when you start examining the the maze of links which intermingle the two major branches. Links are a way of referencing a file or directory by many names and many locations within the file hierarchy. They are effectively like "pointers" to files - think of them as like leaving a post-it note saying "see this file". Links will be explained in greater detail in the next chapter. To put it simply, /var is for VARiable data/files. /usr is for USeR accessible data, programs and libraries. Unfortunately, history has confused things - files which should have been placed in the /usr branch have been located in the
David Jones (20.01.00) Page 88

85321, Systems Administration

Chapter 4: The File Hierarchy

branch and vice versa. Thus to "correct" things, a series of links have been put in place. Why the reason for the separation? Does it matter. The answer is: Yes, but No :)
/var

Yes in the sense that the file standard dictates that the /usr branch should be able to be mounted (another way of saying "attached" to the file hierarchy this will be covered in the next chapter) READ ONLY (thus cant contain variable data). The reasons for this are historical and came about because of something called NFS exporting. NFS exporting is the process of one machine (a server) "exporting" its copy of the /usr structure (and others) to the network for other systems to use. If several systems were "sharing" the same /usr structure, it would not be a good idea for them all to be writing logs and variable data to the same area! It is also used because minimal installations of Linux can use the /usr branch directly from the CDROM (a read-only device). However, it is "No" in the sense that: is usually mounted READ-WRITE-EXECUTE on Linux systems anyway

/usr

In the authors experience, exporting /usr READ-ONLY via NFS isnt entirely successful without making some very non-standard modifications to the file hierarchy!

The following are a few highlights of the /var and /usr directory branches:

/usr/local
All software that is installed on a system after the operating system package itself should be placed in the /usr/local directory. Binary files should be located in the /usr/local/bin (generally /usr/local/bin should be included in a users PATH setting). By placing all installed software in this branch, it makes backups and upgrades of the system far easier - the system administrator can back-up and restore the entire /usr/local system with more ease than backing-up and restoring software packages from multiple branches (i.e.. /usr/src, /usr/bin etc.). An example of a /usr/local directory is listed below:
bin man mpeg netscape games sbin speak src lib volume-1.11 www rsynth info etc cern java

As you can see, there are a few standard directories (bin, lib and src) as well as some that contain installed programs.

David Jones (20.01.00)

Page 89

85321, Systems Administration

Chapter 4: The File Hierarchy

lib, include and src

Linux is a very popular platform for C/C++, Java and Perl program development. As we will discuss in later chapters, Linux also allows the system administrator to actually modify and recompile the kernel. Because of this, compilers, libraries and source directories are treated as "core" elements of the file hierarchy structure. The /usr structure plays host to three important directories: /usr/include holds most of the standard C/C++ header files - this directory will be referred to as the primary include directory in most Makefiles. Makefiles are special script-like files that are processed by the program for the purposes of compiling, linking and building programs.
make

holds most static libraries as well as hosting subdirectories containing libraries for other (non C/C++) languages including Perl and TCL. It also plays host to configuration information for ldconfig.
/usr/lib /usr/src holds the source files for most packages installed on the system. This is traditionally the location for the Linux source directory (/usr/src/linux), for example:
linux linux-2.0.31 redhat

Unlike DOS/Windows based systems, most Linux programs usually come as source and are compiled and installed locally

/var/spool
This directory has the potential for causing a system administrator a bit of trouble as it is used to store (possibly) large volumes of temporary files associated with printing, mail and news. /var/spool may contain something like:
at cron lp mail lpd rwho mqueue uucp samba uucppublic

In this case, there is a printer spool directory called lp (used for storing print request for the printer lp) and a /var/spool/mail directory that contains files for each users incoming mail. Keep an eye on the space consumed by the files and directories found in /var/spool. If a device (like the printer) isn't working or a large volume of e-mail has been sent to the system, then much of the hard drive space can be quickly consumed by files stored in this location.

David Jones (20.01.00)

Page 90

85321, Systems Administration

Chapter 4: The File Hierarchy

X Windows
X-Windows provides UNIX with a very flexible graphical user interface. Tracing the X Windows file hierarchy can be very tedious, especially when your are trying to locate a particular configuration file or trying to removed a stale lock file. A lock file is used to stop more than one instance of a program executing at once, a stale lock is a lock file that was not removed when a program terminated, thus stopping the same program from restarting again Most of X Windows is located in the /usr structure, with some references made to it in the /var structure. Typically, most of the action is in the /usr/X11R6 directory (this is usually an alias or link to another directory depending on the release of X11 - the X Windows manager). This will contain:
bin doc include lib man

The main X Windows binaries are located in /usr/X11R6/bin. This may be accessed via an alias of /usr/bin/X11 . Configuration files for X Windows are located in /usr/X11R6/lib. To really confuse things, the X Windows configuration utility, xf86config, is located in /usr/X11R6/bin, while the configuration file it produces is located in /etc/X11 (XF86Config)! Because of this, it is often very difficult to get an "overall picture" of how X Windows is working - my best advice is read up on it before you start modifying (or developing with) it.

Bins
Which bin?
A very common mistake amongst first time UNIX users is to incorrectly assume that all "bin" directories contain temporary files or files marked for deletion. This misunderstanding comes about because: People associate the word "bin" with rubbish
   

Some unfortunate GUI based operating systems use little icons of "trash cans" for the purposes of storing deleted/temporary files.

However, bin is short for binary - binary or executable files. There are four major bin directories (none of which should be used for storing junk files :)
/bin /sbin /usr/bin /usr/local/bin

Why so many?

David Jones (20.01.00)

Page 91

85321, Systems Administration

Chapter 4: The File Hierarchy

All of the bin directories serve similar but distinct purposes; the division of binary files serves several purposes including ease of backups, administration and logical separation. Note that while most binaries on Linux systems are found in one of these four directories, not all are.
/bin

This directory must be present for the OS to boot. It contains utilities used during the startup; a typical listing would look something like:
Mail arch ash bash cat chgrp chmod domainname-yp chown compress cp cpio false csh cut date dd df dialog dircolors dmesg dnsdomainname domainname ln du echo ed mailx free ftp getoptprog gunzip gzip head hostname ipmask kill killall ps login ls mail rmdir mkdir mkfifo mknod more mount mt mt-GNU mv netstat ping touch pwd red rm umssync setserial setterm sh sln stty su sync tar tcsh telnet

true ttysnoops umount

uname zcat zsh

Note that this directory contains the shells and some basic file and text utilities (ls, pwd, cut, head, tail, ed etc). Ideally, the /bin directory will contain as few files as possible as this makes it easier to take a direct copy for recovery boot/root disks.
/sbin

/sbin Literally "System Binaries". This directory contains files that should generally only be used by the root user, though the Linux file standard dictates that no access restrictions should be placed on normal users to these files. It should be noted that the PATH setting for the root user includes /sbin, while it is (by default) not included in the PATH of normal users. The /sbin directory should contain essential system administration scripts and programs, including those concerned with user management, disk administration, system event control (restart and shutdown programs) and certain networking programs. As a general rule, if users need to run a program, then it should not be located in /sbin. A typical directory listing of /sbin looks like:
adduser agetty arp badblocks bdflush chattr clock debugfs depmod dosfsck dumpe2fs e2fsck explodepkg fdisk ifconfig init insmod installpkg kbdrate killall5 ksyms ldconfig lilo liloconfig liloconfig-color lsattr lsmod makebootdisk mkfs.minix mklost+found mkswap mkxfs modprobe mount netconfig netconfig.color netconfig.tty pidof pkgtool pkgtool.tty plipconfig ramsize rmmod rmt rootflags route runlevel setup setup.tty shutdown swapdev swapoff swapon telinit tune2fs umount

David Jones (20.01.00)

Page 92

85321, Systems Administration

Chapter 4: The File Hierarchy

fsck

makepkg

rarp

fsck.minix genksyms halt

mkdosfs mke2fs mkfs

rdev reboot removepkg

update vidmode xfsck

The very important ldconfig program is also located in /sbin. While not commonly used from the shell prompt, ldconfig is an essential program for the management of dynamic libraries (it is usually executed at boot time). It will often have to be manually run after library (and system) upgrades. You should also be aware of: /usr/sbin - used for non-essential admin tools. /usr/local/sbin - locally installed admin tools.
/usr/bin

This directory contains most of the user binaries - in other words, programs that users will run. It includes standard user applications including editors and email clients as well as compilers, games and various network applications. A listing of this directory will contain some 400 odd files. Users should definitely have /usr/bin in their PATH setting.
/usr/local/bin

To this point, we have examined directories that contain programs that are (in general) part of the actual operating system package. Programs that are installed by the system administrator after that point should be placed in /usr/local/bin. The main reason for doing this is to make it easier to back up installed programs during a system upgrade, or in the worst case, to restore a system after a crash. The /usr/local/bin directory should only contain binaries and scripts - it should not contain subdirectories or configuration files.

Configuration files, logs and other bits!

etc etc etc.
is one place where the root user will spend a lot of time. It is not only the home to the all important passwd file, but contains just about every configuration file for a system (including those for networking, X Windows and the file system). The /etc branch also contains the skel, X11 and rc.d directories. /etc/skel contains the skeleton user files that are placed in a users directory when their account is created. /etc/X11 contains configuration files for X Windows.
/etc

David Jones (20.01.00)

Page 93

85321, Systems Administration

Chapter 4: The File Hierarchy

/etc/rc.d is contains rc directories - each directory is given by the name rcn.d (n is the run level) - each directory may contain multiple files that will

be executed at the particular run level. A sample listing of a /etc/rc.d directory looks something like:
init.d rc rc.local rc.sysinit rc0.d rc1.d rc2.d rc3.d rc4.d rc5.d rc6.d

Logs
Linux maintains a particular area in which to place logs (or files which contain records of events). This directory is /var/log. This directory usually contains:
cron cron.1 cron.2 dmesg httpd lastlog log.nmb log.smb maillog maillog.1 maillog.2 messages messages.1 messages.2 samba samba-log. samba.1 samba.2 secure secure.1 secure.2 sendmail.st spooler spooler.1 spooler.2 uucp wtmp xferlog xferlog.1 xferlog.2

/proc

The /proc directory hierarchy contains files associated with the executing kernel. The files contained in this structure contain information about the state of the systems resource usage (how much memory, swap space and CPU is being used), information about each process and various other useful pieces of information. We will examine this directory structure in more depth in later chapters. The /proc file system is the main source of information for a program called top. This is a very useful administration tool as it displays a "live" readout of the CPU and memory resources being used by each process on the system.
/dev

We will be discussing /dev in detail in the next chapter, however, for the time being, you should be aware that this directory is the primary location for special files called device files.

Conclusion
Future standards
Because Linux is a dynamic OS, there will no doubt be changes to its file system as well. Two current issues that face Linux are: Porting Linux on to may architectures and requiring a common location for hardware independent data files and scripts - the current location is /usr/share - this may change.
 

The location of third-party commercial software on Linux systems - as Linuxs popularity increases, more software developers will produce commercial software to install on Linux systems. For this to happen, a

David Jones (20.01.00)

Page 94

85321, Systems Administration

Chapter 4: The File Hierarchy

location in which this can be installed must be provided and enforced within the file system standard. Currently, /opt is the likely option. Because of this, it is advisable to obtain and read the latest copy of the file system standard so as to be aware of the current issues. Other information sources are easily obtainable by searching the web. You should also be aware that while (in general), the UNIX file hierarchy looks similar from version to version, it contains differences based on requirements and the history of the development of the operating system implementation.

Review Questions
4.1
You have just discovered that the previous system administrator of the system you now manage installed netscap in /sbin. Is this an appropiate location? Why/Why not?.

4.2
Where are man pages kept? Explain the format of the man page directories. (Hint: I didnt explain this anywhere in this chapter - you may have to do some looking)

4.3
As a system administrator, you are going to install the following programs, in each case, state the likely location of each package: Java compiler and libraries
   

DOOM (a loud, violent but extremely entertaining game) A network sniffer (for use by the sys admin only) A new kernel source A X Windows manager binary specially optimised for your new monitor

David Jones (20.01.00)

Page 95

85321, Systems Administration

Chapter 5: Processes and Files

Chapter
Processes and Files

Introduction
This chapter introduces the important and related UNIX concepts of processes and files. A process is basically an executing program. All the work performed by a UNIX system is carried out by processes. The UNIX operating system stores a great deal of information about processes and provides a number of mechanisms by which you can manipulate both the files and the information about them. All the long term information stored on a UNIX system, like most computers today, is stored in files which are organised into a hierarchical directory structure. Each file on a UNIX system has a number of attributes that serve different purposes. As with processes there are a collection of commands which allow users and Systems Administrators to modify these attributes. Among the most important attributes of files and processes examined in this chapter are those associated with user identification and access control. Since UNIX is a multi-user operating system it must provide mechanisms which restrict what and where users (and their processes) can go. An understanding of how this is achieved is essential for a Systems Administrator.

Other Resources
Other resources which discuss some of the concepts mentioned in this chapter include Chapter 17 of this text This is the security chapter of the text and not surprisingly it includes a discussion of file permissions including some additional material which is not discussed here. This chapter is actually a copy of the Security HOWTO from the LDP. Online lecture 5 (which includes slides and audio) Included on the 85321 website/CD-ROM this lecture discusses many of the topics covered in this chapter. You may find it useful to take a listen to this lecture as a supplement to the chapter. Guides on the LDP The Linux Installation and Getting Started Guide has a number of sections looking at the permissions and job control

David Jones (20.01.00)

Page 96

85321, Systems Administration

Chapter 5: Processes and Files

Multiple users
UNIX is a multi-user operating system. This means that at any one time there are multiple people all sharing the computer and its resources. The operating system must have some way of identifying the users and protecting one users resources from the other users.

Identifying users
Before you can use a UNIX computer you must first log in. The login process requires that you have a username and a password. By entering your username you identify yourself to the operating system.

Users and groups

In addition to a unique username UNIX also places every user into at least one group. Groups are used to provide or restrict access to a collection of users and are specified by the /etc/group file. To find out what groups you are a member of use the groups command. It is possible to be a member of more than one group. The following example is taken from my Redhat 6.1 machine
[david@faile links]$ groups david

Executing the groups command as the "normal" user david shows that he is only a member of the david group. Under Linux when you create a user with the adduser command the default action is to create a group with the same name as the account. In the following I user the su command to change to the root user (this requires that I enter roots password). Remember you should do the absolute minimum as root.
[david@faile links]$ su Password: [root@faile /root]# groups root bin daemon sys adm disk wheel

From this you can see that the root user is a member of a number of groups.

Names and numbers

As youve seen each user and group has a unique name. However the operating system does not use these names internally. The names are used for the benefit of the human users. For its own purposes the operating system actually uses numbers to represent each user and group (numbers are more efficient to store). This is achieved by each username having an equivalent user identifier (UID) and every group name having an equivalent group identifier (GID). The association between username and UID is stored in the /etc/passwd file. The association between group name and GID is stored in the /etc/group file.
David Jones (20.01.00) Page 97

85321, Systems Administration

Chapter 5: Processes and Files

To find out the your UID and initial GID try the following command
grep username /etc/passwd

Where username is your username. This command will display your entry in the /etc/passwd file. The third field is your UID and the fourth is your initial GID. On my system my UID is 500 and my GID is 100.
bash$ grep david /etc/passwd david:*:500:100:David Jones:/home/david:/bin/bash id

The id command can be used to discover username, UID, group name and GID of any user.
dinbig:~$ id uid=500(david) gid=100(users) groups=100(users) dinbig:~$ id root uid=0(root) gid=0(root) groups=0(root),1(bin), 2(daemon),3(sys),4(adm),6(disk),10(wheel),11(floppy)

In the above you will see that the user root is a member of more than one group. The entry in the /etc/passwd file stores the GID of the users initial group (mine is 100, roots is 0). If a user belongs to any other groups they are specified in the /etc/group file.

Commands and processes

Whenever you run a program, whether it is by typing in at the command line or running it from X-Windows, a process is created. It is the process, a program in execution and a collection of executable code, data and operating system data structures, which perform the work of the program. The UNIX command line that you use to enter commands is actually another program/command called the shell. The shell is responsible for asking you for a command and then attempting to execute the command. (The shell also performs a number of other tasks which are discussed in the next chapter).

Where are the commands?

For you to execute a command, for example ls, that command must be in one of the directories in your search path. The search path is a list of directories maintained by the shell. When you ask the shell to execute a command it will look in each of the directories in your search path for a file with the same name as the command. When it finds the executable program it will run it. If it doesnt find the executable program it will report command_name: not found.
which

Linux and most UNIX operating systems supply a command called which. The purpose of this command is to search through your search path for a particular command and tell you where it is.

David Jones (20.01.00)

Page 98

85321, Systems Administration

Chapter 5: Processes and Files

For example, the command which ls on my machine aldur returns /usr/bin/ls. This means that the program for ls is in the directory /usr/bin. If you do which for ls on a Redhat Linux machine you will get a different location. Exercises
5.1.

Use the which command to find the locations of the following commands
ls echo set

Why cant I run my shell script?

When you get to chapter 8 of the textbook you will be introduced to shell scripts. Shell scripts are small executable files that contain a bunch of commands, somewhat like batch files under MS-DOS (only better). A common problem many people have when they create their first shell script is that it cant be found. For example, lets assume Ive create a shell script called hello in the current directory. The problem goes something like this.
[david@faile links]$ pwd /home/david/teaching/85321/2000/textbook/mine/links [david@faile links]$ ls -l hello -rwxrwxr-x 1 david david 34 Jan 8 17:15 hello [david@faile links]$ hello bash: hello: command not found

To start with I find out what the current directory is, you will see why in the next couple of paragraphs. I then use the ls command to confirm that the executable file hello is located in the current directory. Then, at last, Ive tried to execute it but get an error message. As mentioned above "command not found" means that the shell was unable to locate the executable file in the current search path. If you think about it you should figure out that this means that the current directory is not in the search path. Thats why the shell cant find the command hello. There are two solutions to this problem Tell the shell exactly the location of the hello executable file. By just typing the name of the command I am telling the shell to search the path. I can be a little more specific with the location using either relative or absolute paths.
[david@faile links]$ /home/david/teaching/85321/2000/textbook/mine/links/hello hello david, how are you [david@faile links]$ ./hello hello david, how are you

Include the current directory in the search path. The idea is to modify the search path so that the shell also looks in the current

David Jones (20.01.00)

Page 99

85321, Systems Administration

Chapter 5: Processes and Files

directory. Absolute and relative paths play a part here also. You will see an explanation of how to change the path in a later chapter.
[david@faile links]$ PATH=$PATH:. [david@faile links]$ hello hello david, how are you

When is a command not a command?

In the previous exercise you will have discovered that which could not find the command. How can this be possible? If I enter the set command on my Linux box it works fine. So if all commands are executable files in the search path then why cant which find it? This is because set is a built-in shell command. This means there isnt an executable program that contains the code for the set command. Instead the code for set is actually built into the shell. In other words no matter how hard you look you wont find an executable file called set. So, as mentioned before any command you execute at a UNIX command line falls into one of two categories A shell command. This is a command which is understood by the shell you are using. It isnt an executable file.
set

An executable file. The executable file will be located somewhere in your search path. When you execute this type of command the shell will search for the file and then create a process which executes this file.

Why shell commands are faster than other commands

As mentioned above executing a shell command does not require the creation of a new process, the existing shell process executes the command. For normal commands a new process must be created. Creating a new process is, relatively speaking, quite a long process. This is especially true when the executable file must be read from disk (you should remember from operating systems that reading from disk is very, very slow when compared to RAM and CPU operations). This is why internal shell commands are much faster than normal commands. For example, I have created two shell scripts (add and add2) which both perform the simple task of adding up to 1000 1 at a time. add uses a normal command to perform the addition. While add2 uses an internal shell command to perform the addition. To compare the speed of the two scripts I use the UNIX time command to work out how long each script takes to execute
[david@faile links]$ time add 6.82user 7.15system 0:13.97elapsed 99%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (107194major+70036minor)pagefaults 0swaps [david@faile links]$ time add2 0.52user 0.00system 0:00.51elapsed 100%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (194major+24minor)pagefaults 0swaps

David Jones (20.01.00)

Page 100

85321, Systems Administration

Chapter 5: Processes and Files

From the output of these two commands you should be able to see that using the internal shell command is significantly faster than using the normal UNIX command. The drawback of shell commands is that they can only be used with a specific shell, you might not be using the right shell. On the other hand, the common UNIX commands are present on all UNIX systems.

Controlling processes
Processes are the main active component of any modern operating system. Any work performed by a modern operating system is performed by processes. UNIX/Linux is no different. This section provides an overview of how you can view and manipulate processes as a normal user. This is a primary responsibility for a Systems Administrator so it is important foundation knowledge. In this section you will learn how to view existing processes Discover how to find out which processes exist, what is their current state and who they belong to. Job control How you can control the execution of processes using the features of common shells. Process manipulation How processes can be stopped or restarted by sending signals.

Online lecture 5 also takes a look at this material.

Viewing existing processes

As mentioned earlier in this chapter every UNIX command you execute runs as a new process. Since Linux/UNIX is a multi-tasking operating system at any one time there can be tens, hundreds even thousands of processes running (the limit is set by a value in the source code for the Linux kernel). As a Systems Administrator and a normal user you will want to be able to find out which processes are currently running, what there current state is and a bunch of other process related information. This section introduces you to a number of commands which allow you to do this including
ps

Provides a snapshot of the processes which are currently running.

top

Provides a full screen, updated view of the current processes.

pstree

Displays a tree-like structure of the current processes. Various graphical tools It is now common for a range of GUI tools to be available. This section will look briefly at those which come with the GNOME desktop environment.
Page 101

David Jones (20.01.00)

85321, Systems Administration

Chapter 5: Processes and Files

ps

Any use on a UNIX system can execute the ps command and see something like
[david@faile linux]$ ps PID TTY TIME CMD 667 pts/0 00:00:00 bash 893 pts/0 00:00:00 ps

This is simply the list of all processes running from the current terminal (TTY which is currently pts/0). The ps command understands a wide range of command-line switches which will modify both the rows, and By modifiying the rows which appear you are changing which processes are shown. By default you are only seeing the processes for the current terminal. The example below shows how this can be changed. columns The columns display various bits of information about the processes. By default you see such things as the commands used the process is running (the CMD column) and the unique process identifier for the process (the PID column).

For example,
[david@faile linux]$ ps a PID TTY STAT TIME 667 pts/0 S 0:00 902 pts/0 R 0:00 [david@faile linux]$ ps x PID TTY STAT TIME 592 tty1 SW 0:00 603 tty1 SW 0:00 610 tty1 SW 0:00 615 tty1 S 0:00 COMMAND bash ps a COMMAND [bash] [startx] [xinit] /usr/bin/gnome-session

..... some output deleted here... 667 pts/0 669 tty1 670 pts/1 671 tty1 672 pts/2 675 tty1 676 tty1 677 tty1 678 pts/3 679 pts/4 680 pts/5 688 tty1 707 tty1 session 720 tty1 721 tty1 722 tty1 723 tty1 724 tty1 725 tty1 727 tty1 795 pts/3 835 tty1 924 pts/0 S SW SW SW SW SW SW SW S S SW S S S S S S S S S S S R 0:00 0:00 0:00 0:00 0:00 0:00 0:00 0:00 0:00 0:00 0:00 1:42 0:41 0:00 0:00 0:00 0:00 0:00 0:00 0:00 0:00 0:26 0:00 bash [gnome-pty-helpe] [bash] [gnome-pty-helpe] [bash] [gnome-pty-helpe] [gnome-pty-helpe] [gnome-pty-helpe] bash bash [bash] /home/david/Office51/bin/soffice.bin /usr/lib/netscape/netscape-communicator -irix(dns helper) /home/david/Office51/bin/soffice.bin /home/david/Office51/bin/soffice.bin /home/david/Office51/bin/soffice.bin /home/david/Office51/bin/soffice.bin /home/david/Office51/bin/soffice.bin /home/david/Office51/bin/soffice.bin vi TODO gtop ps x

David Jones (20.01.00)

Page 102

85321, Systems Administration

Chapter 5: Processes and Files

Refer to the manual page for the ps command for more information about the available switches. You will notice that ps does not follow the standard UNIX command format. In this case the command-line switches a and x were not preceded with -.
5.2.

Use the ps command to discover which user owns the

/usr/sbin/atd sendmail

processes.
top

ps provides a one-off snapshot of the current processes. If you want an ongoing view of the processes you need to use top. top produces output something like
10:56am up 1:21, 7 users, load average: 1.32, 0.80, 0.41 95 processes: 92 sleeping, 2 running, 1 zombie, 0 stopped CPU states: 15.0% user, 3.6% system, 0.0% nice, 81.3% idle Mem: 127948K av, 124496K used, 3452K free, 58884K shrd, 2888K buff Swap: 72252K av, 21956K used, 50296K free 34528K cached PID USER 974 david 977 david 835 david 611 root 1 root 2 root 3 root 4 root 5 root 6 root mdrecoveryd 253 bin 269 root 322 root 333 root 349 daemon 365 root 380 root PRI NI SIZE RSS SHARE STAT 12 0 428 428 348 R 7 0 1044 1044 816 R 1 0 3912 3912 2876 S 0 0 34468 32M 1620 S 0 0 124 72 52 S 0 0 0 0 0 SW 0 0 0 0 0 SW 0 0 0 0 0 SW 0 0 0 0 0 SW -20 -20 0 0 0 SW< 0 0 0 0 0 0 0 0 0 0 0 0 0 0 80 380 200 504 144 240 140 0 368 148 160 104 188 0 0 324 104 112 76 144 0 SW S S S S S SW LIB %CPU %MEM 0 79.0 0.3 0 10.6 0.8 0 9.7 3.0 0 0.8 25.8 0 0.0 0.0 0 0.0 0.0 0 0.0 0.0 0 0.0 0.0 0 0.0 0.0 0 0.0 0.0 0 0 0 0 0 0 0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.2 0.1 0.1 0.0 0.1 0.0 TIME 4:00 0:00 1:55 3:00 0:04 0:00 0:00 0:00 0:00 0:00 0:00 0:00 0:00 0:00 0:00 0:00 0:00 COMMAND yes top gtop X init kflushd kupdate kpiod kswapd

portmap apmd syslogd klogd atd crond cardmgr

As with ps there are a number of command-line switches which modify the operation of top. Additionally top has a number of interactive commands you can use while it is running. For example, hitting the h key while top is running will display a simple help screen which lists the interactive commands.
pstree and ps f

Each new process (the child process) must be started by another process (the parent process). As a result UNIX processes form a family tree. The pstree and the f switch of the ps command allow you to view this family tree. For example
[david@faile david]$ pstree init-+-apmd |-atd |-cardmgr |-crond |-enlightenment |-gen_util_applet

David Jones (20.01.00)

Page 103

85321, Systems Administration

Chapter 5: Processes and Files

|-gmc |-gnome-name-serv |-gnome-smproxy |-gnome-terminal-+-bash---pstree | -gnome-pty-helpe |-3*[gnome-terminal-+-bash] | -gnome-pty-helpe] |-gnome-terminal-+-bash-+-top | | -yes | -gnome-pty-helpe |-gnome-terminal-+-bash---su---bash | -gnome-pty-helpe |-gnomepager_appl |-gpm |-gtop |-httpd---15*[httpd] |-inetd |-kflushd |-klogd |-kpiod |-kswapd |-kupdate |-login---bash---startx---xinit-+-X | -gnome-session |-lpd |-magicdev |-mdrecoveryd |-5*[mingetty] |-msql2d |-netscape-commun---2*[netscape-commun] |-panel |-portmap |-safe_mysqld---mysqld---mysqld---mysqld |-soffice.bin---soffice.bin---5*[soffice.bin] |-syslogd |-xfs -xscreensaver

gtop

The increasing use of X Windows and GUI environments means that there have been a number of GUI tools written to provide similar features as the text-based tools introduced in the previous couple of sections. One of them is gtop, the GNU system monitor program, which by default provides a display not unlike top (but as GUI). Gtop also provides a number of additional services including displays of memory and file system usage. Diagram 5.1 is a screen shot of the memory usage screen.

David Jones (20.01.00)

Page 104

85321, Systems Administration

Chapter 5: Processes and Files

Diagram 1.1 Screen shot of gtop

Job control
Jobs and processes are the same thing in UNIX terminology. Job control is a facility provided by most shells which allow you to switch between multiple running processes. So far most of you will have been running only a single job. Not unlike what was done in the previous examples when running the ps command. The normal process goes something like this You type a command at the shell prompt The shell runs that command while you wait for it complete. When it is finished the shell displays another command-line and you can start again.

During this process the shell goes "to sleep" waiting for the command to finish. You can see this in the ps a example from above. In this example bash is the shell and ps is the command which is being executed. Take a look at the STAT column for bash, it is S. STAT or status indicates the current status for a process. NewTable 5.1 summarises the possible states for a Linux process. This table is adapted from the manual page for the ps command.

David Jones (20.01.00)

Page 105

85321, Systems Administration

Chapter 5: Processes and Files

Process D R S T Z

State codes uninterruptible sleep (usually IO) runnable (on run queue) Sleeping traced or stopped a defunct ("zombie") process
NewTable 5.1 Linux Process States

As you should remember from operating systems on a single CPU system there can only ever be one running process. In the ps a example from above the running process is the one executing the ps command. This running process is called the foreground process (job). It is the process which "owns" the terminal for input and output. Usually there is only one running process. However most shells provide mechanisms by which you can interrupt a process Interrupting a process is the same as killing it. The process dies i.e. is no longer running. The typical method for interrupting the current foreground process is using the CTRL-C key combination (hold the control key down and hit the c key). For example, run the yes command which continues to display a line of ys one to a line. The yes command will quite happily do this forever. To stop it hit CTRL-C. You have just interrupted a process. suspend a process Suspending a process puts it to sleep until you start it again. You use the key combination CTRL-Z to suspend a process. Run the yes command again. This time suspend it rather than interrupt it. You should see something like

y y [1]+ Stopped yes

The [1] is the job number for the suspended process. You can use this to restart the process. If you now run the ps a command you will see something like
[david@faile 2000]$ ps a PID TTY STAT TIME COMMAND 678 pts/3 S 0:00 bash 962 pts/3 T 0:00 yes 963 pts/3 R 0:00 ps a

Notice that the yes process appears, so it still exists. If you refer back to the previous table you can see that its state is now stopped. check on the status of jobs The jobs command is used to check on the status of the jobs you currently

David Jones (20.01.00)

Page 106

85321, Systems Administration

Chapter 5: Processes and Files

have associated with the terminal. In our current situation you get something like
[david@faile 2000]$ jobs [1]+ Stopped yes

change the current foreground process To put the yes command back into the foreground (to take it out of the background) you can use the fg command. fg %1 will put the yes command back into the foreground and start the ys scrolling down the screen again. The %1 is used to indicate which job you want back into the foreground. The 1 matches the [1] displayed when we stopped the job above. Feel free to interrupt the job at any stage. run other processes in the background The shells also support the idea of starting a process off in the background. This means that the command you execute goes straight into the background rather than staying in the foreground. This is achieved using the & symbol. For example

[david@faile 2000]$ yes > /dev/null & [1] 974 [david@faile 2000]$ jobs [1]+ Running yes >/dev/null & [david@faile 2000]$ ps a PID TTY STAT TIME COMMAND 678 pts/3 S 0:00 bash 974 pts/3 R 0:35 yes 976 pts/3 R 0:00 ps a

The [1] 974 indicates that the yes command has become job number 1 with process id 974. This is reinforced by using the jobs and ps a commands to view the current jobs and processes. Notice that we now have two processes which are on the runnable queue, ps and yes.

Manipulating processes
You have already seen some simple approaches to manipulating processes using the CTRL-C and CTRL-Z key combinations. These approaches along with all approaches to manipulating processes are related to sending signals to processes. When a process is executed it automatically has a collection of signal handlers create. Each signal handler is essentially a function which is executed when a certain signal is received. If you are interested in finding out more about signals you can refer to online lecture 5 or to the manual page signal(7). This manual page describes all 30 signals used by Linux and also the default actions which are expected as a result of receiving a particular signal. The kill command Apart from using certain key combinations you can also send signals to processes using the kill command. The kill command is used to send a specific signal to a specific process. This means you usually have to specify both the signal and the process.

David Jones (20.01.00)

Page 107

85321, Systems Administration

Chapter 5: Processes and Files

By default the kill command sends the TERM signal. You can specify other signals by using the appropriate number of title. The -l switch of the kill command provides a quick overview of the available signals, their names and numbers.
[david@faile david]$ kill -l 1) SIGHUP 2) SIGINT 3) SIGQUIT 4) SIGILL 5) SIGTRAP 6) SIGIOT 7) SIGBUS 8) SIGFPE 9) SIGKILL 10) SIGUSR1 11) SIGSEGV 12) SIGUSR2 13) SIGPIPE 14) SIGALRM 15) SIGTERM 17) SIGCHLD 18) SIGCONT 19) SIGSTOP 20) SIGTSTP 21) SIGTTIN 22) SIGTTOU 23) SIGURG 24) SIGXCPU 25) SIGXFSZ 26) SIGVTALRM 27) SIGPROF 28) SIGWINCH 29) SIGIO 30) SIGPWR

You specify the process to which you want to send a signal using the process identifier as shown by the ps or top commands. The following commands demonstrate how job control, the ps command and the kill command can be combined.
[david@faile [2] 1187 [1] Killed [david@faile [3] 1188 [david@faile [4] 1189 [david@faile PID TTY 678 pts/3 1187 pts/3 1188 pts/3 1189 pts/3 1190 pts/3 2000]$ yes > /dev/null & yes >/dev/null 2000]$ yes > /dev/null & 2000]$ yes > /dev/null & 2000]$ ps a STAT TIME COMMAND S 0:00 bash R 0:13 yes R 0:11 yes R 0:11 yes R 0:00 ps a

To start with we create three versions of the yes command all running in the background. We know start sending some signals to the processes using the kill command. In the first kill command I dont specify a signal. This means the kill command will use the default TERM signal. The names of signals are shown in the kill -l output from above. However, you wont see a name TERM, you will see the name SIGTERM. When used in the kill command and in some discussions the SIG is dropped. So the KILL signal is called SIGKILL above.
[david@faile 2000]$ kill 1187 [david@faile 2000]$ ps a PID TTY STAT TIME COMMAND 678 pts/3 S 0:00 bash 1188 pts/3 R 0:40 yes 1189 pts/3 R 0:39 yes 1193 pts/3 R 0:00 ps a [2] Terminated yes >/dev/null

From the message and the output of the ps command you can see that process 1187 has been destroyed.
[david@faile 2000]$ kill -STOP 1188 [3]+ Stopped (signal) yes >/dev/null [david@faile 2000]$ kill -19 1189

David Jones (20.01.00)

Page 108

85321, Systems Administration

Chapter 5: Processes and Files

[david@faile 2000]$ [4]+ Stopped (signal) [david@faile 2000]$ ps a PID TTY STAT TIME 678 pts/3 S 0:00 1188 pts/3 T 0:53 1189 pts/3 T 1:11 1195 pts/3 R 0:00 yes >/dev/null COMMAND bash yes yes ps a

In the previous commands the two processes 1188 and 1189 have been suspended using the kill command instead of using the CTRL-Z key combination. This demonstrates that when you use the CTRL-Z key combination you are actually sending the process the SIGSTOP (signal number 19) signal.
[david@faile 2000]$ kill -5 1188 [david@faile 2000]$ ps a PID TTY STAT TIME COMMAND 678 pts/3 S 0:00 bash 1188 pts/3 T 0:53 yes 1189 pts/3 T 1:11 yes 1200 pts/3 R 0:00 ps a

From these commands it appears that sending signal 5 (SIGTRAP) has no effect on process 1188. Exercises
5.3.

Under the VMS operating system it is common to use the key combination CTRL-Z to kill a program. A new user on your UNIX system has been using VMS a lot. What happens when he uses CTRL-Z while editing a document with vi?

Process attributes
For every process that is created the UNIX operating system stores information including its real UID, GID and its effective UID and GID These are used to identify the owner of the process (real UID and GID) and determine what the process is allowed to do (effective UID and GID) the code and variables used by the process (its address map) the status of the process its priority its parent process

Parent processes
All processes are created by another process (its parent). The creation of a child process is usually a combination of two operations

David Jones (20.01.00)

Page 109

85321, Systems Administration

Chapter 5: Processes and Files

forking A new process is created that is almost identical to the parent process. It will be using the same code. exec This changes the code being used by the process to that of another program.

When you enter a command it is the shell that performs these tasks. It will fork off a new process (which is running the shells program). The child process then performs an exec to change to the code for the command you wish executed. Examples of this are shown in the pstree section earlier in this chapter.

Process UID and GID

In order for the operating system to know what a process is allowed to do it must store information about who owns the process (UID and GID). The UNIX operating system stores two types of UID and two types of GID.

Real UID and GID

A process real UID and GID will be the same as the UID and GID of the user who ran the process. Therefore any process you execute will have your UID and GID. The real UID and GID are used for accounting purposes.

Effective UID and GID

The effective UID and GID are used to determine what operations a process can perform. In most cases the effective UID and GID will be the same as the real UID and GID. However using special file permissions it is possible to change the effective UID and GID. How and why you would want to do this is examined later in this chapter. The following exercise asks you to create an executable program we will use to display the real and effective UID and GID.

David Jones (20.01.00)

Page 110

85321, Systems Administration

Chapter 5: Processes and Files

Exercises
5.4.

Create a text file called i_am.c that contains the following C program. Compile the program by using the following command
cc i_am.cc -o i_am

This will produce an executable program called i_am. Run the program. (rather than type the code, you should be able to cut and paste it from the online versions of this chapter that are on the CD-ROM and Web site)
#include <stdio.h> #include <unistd.h> void main() { int real_uid, effective_uid; int real_gid, effective_gid; /* get the user id and group id*/ real_uid = getuid(); effective_uid = geteuid(); real_gid = getgid(); effective_gid = getegid(); /* display what I found */ printf( "The real uid is %d\n", real_uid ); printf("The effective uid is %d\n", effective_uid ); printf("The real gid is %d\n", real_gid ); printf("The effective gid is %d\n", effective_gid ); }
5.5.

Make sure you are logged in as a normal user when you start the following exercise. In a previous exercise you were asked to discover which user owns the /usr/sbin/atd and sendmail processes. Try to cause these programs to stop using the kill command. If it doesnt work, why not? There are two reasons which may explain this problem. What are they? Use the ps command to discover which user is the "owner" of the atd and sendmail processes

5.6.

Files
All the information stored by UNIX onto disk is stored in files. Under UNIX even directories are just special types of files. A previous reading has already introduced you to the basic UNIX directory hierarchy. The purpose of this section is to fill in some of the detail including discussion of file types UNIX recognises a number of special file types which are used for specific purposes (e.g. a directory is a special file type). Normal files Normal files, those used to store data, can also have a number of types which describe the type of data stored in the file (e.g. a GIF file, a Word document) file attributes The file type is just one of the attributes UNIX stores about files. There are many others including owner and size.

David Jones (20.01.00)

Page 111

85321, Systems Administration

Chapter 5: Processes and Files

File types
UNIX supports a small number of different file types. The following table summarises these different file types. What the different file types are and what their purpose is will be explained as we progress. File types are signified by a single character which is used in the output of the ls command (you use the ls command to view the various attributes of a file).. File type
d l b c p

Meaning a normal file a directory symbolic link block device file character device file a fifo or named pipe
Table 5.1 UNIX file types

For current purposes you can think of these file types as falling into three categories normal files, Normal files are used to store data and under UNIX they are just a collection of bytes of information. The format of these bytes can be used to identify a normal file as a GIF file or a Word document. directories or directory files, Remember, for UNIX a directory is just another file which happens to contain the names of files and their I-node. An I-node is an operating system data structure which is used to store information about the file (explained later). special or device files. Explained in more detail later on in the text these special files provide access to devices which are connected to the computer. Why these exist and what they are used for will be explained.

Types of normal files

Those of you who use Windows will be familiar with normal files having different types (e.g. GIF images, Word documents etc). Under Windows the type of a normal file is specified by its extension. UNIX does not use this approach. In fact the operating system makes no distinction between different types of files. All files are simply a collection of bytes. However, UNIX does provide commands which allow you to determine the type of normal files. If you're unsure what type of normal file you have the UNIX file command might help.

David Jones (20.01.00)

Page 112

85321, Systems Administration

Chapter 5: Processes and Files

[david@faile david]$ file article.doc reopen.call gtop.gif pair.pdf /etc/passwd article.doc: Microsoft Office Document reopen.call: Microsoft Office Document gtop.gif: GIF image data, version 89a, 618 x 428, pair.pdf: PDF document, version 1.2 /etc/passwd: ASCII text

In this example the file command has been used to discover what type of file for a number of files. Some important things to notice extension doesnt matter The file reopen.call is a Word document but its extension is not .doc. Additional features For some file types the file command provides additional features such as the height and width of the GIF image and the version of PDF used in the PDF file.

How does the file command work? The file command attempts to perform three tests on a file to determine its type. The first test which works is used. The three tests are file system tests This works if the file to be tested is one of the special files listed in the previous section (e.g. a directory, device file etc). For example
[david@faile 2000]$ file /home /dev/hda /home: directory /dev/hda: block special (3/0)

magic number tests Many data file formats always contain a specific value at a specific location in the file. These value is referred to as a magic number. UNIX systems maintain a data file (/usr/share/magic on Linux) which contains a collection of magic numbers for various file types (take a look at the file on your Linux computer). language tests Finally if the file is a text file it attempts to determine what type of computer language the file contains. Exercises
5.7.

Examine the contents of the /usr/share/magic file. Experiment with the file command on a number of different files.

File attributes
The UNIX/Linux operating system uses a data structure called an inode to store all of the information it stores about a file (except for the filename). Every file on a UNIX system must have an associated inode on the disk. If you run out of inodes you cant create any more files on that disk. You can find out which inode a file has by using the ls -i command.
dinbig:~$ ls -i README 45210 README

David Jones (20.01.00)

Page 113

85321, Systems Administration

Chapter 5: Processes and Files

In the above example the file README is using inode 45210. Some fo the information UNIX stores about each file includes where the files data is stored on the disk This is the collection of disk blocks which are used to store the data for the file. what the files name is The name of a file is actually stored in the directory in which it occurs. Each entry in a directory contains a filename and an associated inode number. who owns the file The inode contains the UID and GID of the user who owns the file. who is allowed to do what with the file This is stored in the file permissions of a file. We examine file permissions in more detail below. how big the file is when was the file last modified how many links there are to the file It is possible for the same file to be known by more than one name. Remember the filename is stored in a directory. It is possible to have many different directories contain pointers to the one inode.

Throughout this text you will find the term file used to mean both files and directories.

Viewing file attributes

To examine the various attributes associated with a file you can use the -l switch of the ls command. This section provides some additional information about the file attributes.

Figure 5.1 File Attributes

Filenames Most UNIX file systems (including the Linux file system) will allow filenames to be 255 characters long and use almost any characters. However there are

David Jones (20.01.00)

Page 114

85321, Systems Administration

Chapter 5: Processes and Files

some characters that can cause problems if used including * $ ? " / \ and others (including the space character). Why is explained in the next chapter. This doesnt mean you cant create filenames that contain these characters, just that you can have some problems if you do. Size The size of a file is specified in bytes. So the above file is 227 bytes long. The standard Linux file system, called EXT2, will allow files to be up to 2Gb (giga bytes) in size. Which these days is not that large. Development is currently ongoing for EXT3 which will increase this file size. Owner and Group Owner Even though the ls command displays the names of the user and group owner of a file that is not what is stored on the inode. The main reason being is that it would consume too much space to store the names. Instead the inode contains the UID and GID of the user and group owner. The ls command performs a translation from UID/GID to the name when it executes. Date The date specified here is the date the file was last modified. Permissions The permission attributes of a file specifies what operations can be done with a file and who can perform those operations. Permissions are explained in more detail in the following section. Exercises
5.8.

Examine the following command and its output (executing these commands on your system should provide very similar results).
[david@faile 3]$ ls -ld / /dev drwxr-xr-x 19 root root 1024 Dec 6 11:30 / drwxr-xr-x 2 root root 22528 Dec 8 10:12 /dev

Answer the following questions 1. What type of files are / and /dev? 2. What else can you tell about these files? 3. How come /dev is bigger than /?

David Jones (20.01.00)

Page 115

85321, Systems Administration

Chapter 5: Processes and Files

5.9.

Execute the following commands mkdir tmp ls -ld tmp touch tmp/tempfiledj ls -ld tmp These commands create a new directory called tmp and create an empty file tempfiledj inside that directory. The touch command is used to create an empty file if the file doesnt exist, or updates the date last modified if it does. Why does the output of the ls -ld tmp command change?

File protection
Given that there can be many people sharing a UNIX computer it is important that the operating system provide some method of restricting access to files. I dont want you to be able to look at my personal files. UNIX achieves this by restricting users to three valid operations, Under UNIX there are only three things you can do to a file (or directory): read, write or execute it. allow the file owner to specify who can do these operations on a file. The file owner can use the user and group concepts of UNIX to restrict which users (actually it restricts which processes that are owned by particular users) can perform these tasks.

File operations
UNIX provides three basic operations that can be performed on a file or a directory. The following table summarises those operations. It is important to recognise that the operations are slightly different depending whether they are being applied to a file or a directory. Operation Effect on a file Effect on a directory read write execute read the contents of the file find out what files are in the directory, e.g. ls delete the file or add something to the file be able to create or remove a file from the directory

be able to run a file/program be able to access a file within a directory

Table 5.2 UNIX file operations

Users, groups and others

Processes wishing to access a file on a UNIX computer are placed into one of three categories
David Jones (20.01.00) Page 116

85321, Systems Administration

Chapter 5: Processes and Files

user The individual user who owns the file (by default the user that created the file but this can be changed). In figure 5.1 the owner is the user david. group The collection of people that belong to the group that owns the file (by default the group to which the files creator belongs). In figure 5.1 the group is staff. other Anybody that doesnt fall into the first two categories.

Figure File

5.2

Permissions

File permissons Each user category (user, group and other) have their own set of file permissions. These control what file operation each particular user category can perform. File permissions are the first field of file attributes to appear in the output of ls -l. File permissions actually consist of four fields file type, user permissions, group permissions, and other permissions.

Three sets of file permissions

As Figure 5.2 shows the file permissions for a file are divided into three different sets one for the user, one for a group which owns the file and one for everyone else. A letter indicates that the particular category of user has permission to perform that operation on the file. A - indicates that they cant. In the above diagram the owner can read, write and execute the file (rwx). The group can read and write the file (rw-), while other cannot do anything with the file (---).
David Jones (20.01.00) Page 117

85321, Systems Administration

Chapter 5: Processes and Files

Symbolic and numeric permissions

rwxr-x-w- is referred to as symbolic permissions. The permissions are represented using a variety of symbols. There is another method for representing file permissions called numeric or absolute permissions where the file permissions are represented using numbers. The relationship between symbolic and numeric permissions is discussed in a couple of pages.

Symbols The following table summarises the symbols that can be used in representing file permissions using the symbolic method. Symbol Purpose
r w x s t

read write execute setuid or setgid (depending on location) sticky bit

Table 5.3 Symbolic file permissions

Special permissions
Table 5.3 introduced three new types of permission setuid, setgid and the sticky bit. Sticky bit on a file In the past having the sticky bit set on a file meant that when the file was executed the code for the program would "stick" in RAM. Normally once a program has finished its code was taken out of RAM and that area used for something else. The sticky bit was used on programs that were executed regularly. If the code for a program is already in RAM the program will start much quicker because the code doesnt have to be loaded from disk. However today with the advent of shared libraries and cheap RAM most modern Unices ignore the sticky bit when it is set on a file. Sticky bit on a directory The /tmp directory on UNIX is used by a number of programs to store temporary files regardless of the user. For example when you use elm (a UNIX mail program) to send a mail message, while you are editing the message it will be stored as a file in the /tmp directory.

David Jones (20.01.00)

Page 118

85321, Systems Administration

Chapter 5: Processes and Files

Modern UNIX operating systems (including Linux) use the sticky bit on a directory to make /tmp directories more secure. Try the command ls -ld /tmp what do you notice about the file permissions of /tmp. If the sticky bit is set on a directory you can only delete or rename a file in that directory if you are the owner of the directory, the owner of the file, or the super user

Effective UID and GID

In this section we revisit the discussion of the relationship between the process attributes of real UID/GID and effective UID/GID. When you use the passwd command to change your password the command will actually change the contents of either the /etc/passwd or /etc/shadow files. These are the files where your password is stored. By default most Linux systems use /etc/passwd As has been mentioned previously the UNIX operating system uses the effective UID and GID of a process to decide whether or not that process can modify a file. Also the effective UID and GID are normally the UID and GID of the user who executes the process. This means that if I use the passwd command to modify the contents of the /etc/passwd file (I write to the file) then I must have write permission on the /etc/passwd file. Lets find out. What are the file permissions on the /etc/passwd file?
dinbig:~$ ls -l /etc/passwd -rw-r--r-1 root root 697 Feb 1 21:21 /etc/passwd

On the basis of these permissions should I be able to write to the /etc/passwd file? No. Only the user who owns the file, root, has write permission. Then how do does the passwd command change my password?

setuid and setgid

This is where the setuid and setgid file permissions enter the picture. Lets have a look at the permissions for the passwd command (first we find out where it is).
dinbig:~$ which passwd /usr/bin/passwd dinbig:~$ ls -l /usr/bin/passwd -rws--x--x 1 root bin /usr/bin/passwd

7192 Oct 16 06:10

Notice the s symbol in the file permissions of the passwd command, this specifies that this command is setuid.

David Jones (20.01.00)

Page 119

85321, Systems Administration

Chapter 5: Processes and Files

The setuid and setgid permissions are used to change the effective UID and GID of a process. When I execute the passwd command a new process is created. The real UID and GID of this process will match my UID and GID. However the effective UID and GID (the values used to check file permissions) will be set to that of the command. In the case of the passwd command the effective UID will be that of root because the setuid permission is set, while the effective GID will be my groups because the setgid bit is not set. Exercises
5.10.

Log in as the root user, go to the directory that contains the file i_am you created in exercise 5.3. Execute the following commands cp i_am i_am_root cp i_am i_am_root_group chown root.root i_am_root* chmod a+rx i_am* chmod u+s i_am_root chmod +s i_am_root_group ls -l i_am* These commands make copies of the i_am program called i_am_root with setuid set, and i_am_root_group with setuid and setgid set. Log back in as your normal user and execute all three of the i_am programs. What do you notice? What is the UID and gid of root?

Numeric permissions
Up until now we have been using symbols like r w x s t to represent file permissions. However the operating system itself doesnt use symbols, instead it uses numbers. When you use symbolic permissions, the commands translate between the symbolic permission and the numeric permission. With numeric or absolute permissions the file permissions are represented using octal (base 8) numbers rather than symbols. The following table summarises the relationship between the symbols used in symbolic permissions and the numbers used in numeric permissions. To obtain the numeric permissions for a file you add the numbers for all the permissions that are allowed together. Symbol s
t r w x

Number 4000 setuid 2000 setgid 1000 400 user 40 group 4 other 200 user 20 group 2 other 100 user 10 group 1 other
Table 5.4 file permissions

Numeric

David Jones (20.01.00)

Page 120

85321, Systems Administration

Chapter 5: Processes and Files

Symbolic to numeric
Heres an example of converting from symbolic to numeric using a different method. This method relies on using binary numbers to calculate the numeric permissions. The process goes something like this write down the symbolic permissions, under each permission that is on, write a one under each permission that is off, write a zero for each category of user, user, group and other convert the three binary digits into decimal, e.g. rwx -> 111 -> 7 combine the three numbers (one each for user, group and other) into a single octal number

Symbolic

to

Figure 5.3 Numeric permissions

Exercises
5.11.

Convert the following symbolic permissions to numeric

rwxrwxrwx -----------r--r-r-sr-x--rwsrwsrwt

5.12.

Convert the following numeric permissions to symbolic

710 4755 5755 6750 7000

David Jones (20.01.00)

Page 121

85321, Systems Administration

Chapter 5: Processes and Files

Changing file permissions

The UNIX operating system provides a number of commands for users to change the permissions associated with a file. The following table provides a summary. Command Purpose
chmod

change the file permissions for a file set the default file permissions for any files to be created. Usually run as the user logs in. change the group owner of a file change the user owner of a file.
change Table 5.5 file ownership and permissions

umask

chgrp chown
Commands to

chmod
The chmod command is used to the change a files permissions. Only the user who owns the file can change the permissions of a file (the root user can also do it). Format
chmod [-R] operation files

The optional (the [ ] are used to indicate optional) switch -R causes chmod to recursively descend any directories changing file permissions as it goes. files is the list of files and directories to change the permissions of. operation indicates how to change the permissions of the files. operation can be specified using either symbolic or absolute permissions. Numeric permissions When using numeric permissions operation is the numeric permissions to change the files permissions to. For example
chmod 770 my.file

will change the file permissions of the file my.file to the numeric permissions
770.

Symbolic permissions When using symbolic permissions operation has three parts who op symbolic_permission where who specifies the category of user to change the permissions for It can be any combination of u for user, g for group, o for others and a for all categories.

David Jones (20.01.00)

Page 122

85321, Systems Administration

Chapter 5: Processes and Files

op specifies how to change the permissions + add permission, - remove permission, = set

permission set sticky bit.

permission specifies the symbolic permissions r for read, w for write, x execute, s set uid/gid, t

Examples
chmod u+rwx temp.dat

add rwx permission for the owner of the file, these permissions are added to the existing permissions
chmod go-rwx temp.dat

remove all permissions for the group and other categories

chmod -R a-rwx /etc

turn off all permissions, for all users, for all files in the /etc directory.
chmod -R a= /

turn off all permissions for everyone for all files

chmod 770 temp.dat

allow the user and group read, write and execute and others no access

chown
The UNIX operating system provides the chown command so that the owner of a file can be changed. However in most Unices only the root user can use the command. Two reasons why this is so are in a file system with quotas (quotas place an upper limit of how many files and how much disk space a user can use) a person could avoid the quota system by giving away the ownership to another person if anyone can give ownership of a file to root they could create a program that is setuid to the owner of the file and then change the owner of the file to root

chgrp

UNIX also supplies the command chgrp to change the group owner of a file. Any user can use the chgrp command to change any file they are the owner of. However you can only change the group owner of a file to a group to which you belong. For example
dinbig$ whoami david dinbig$ groups users dinbig$ ls -l tmp -rwxr-xr-x 2 david users 1024 Feb 1 21:49 tmp dinbig$ ls -l /etc/passwd

David Jones (20.01.00)

Page 123

85321, Systems Administration

Chapter 5: Processes and Files

dinbig$ chgrp users /etc/passwd chgrp: /etc/passwd: Operation not permitted -rw-r--r-- 1 root root 697 Feb 1 21:21 /etc/passwd dinbig$ chgrp man tmp chgrp: you are not a member of group man: Operation not permitted

In this example Ive tried to change the group owner of /etc/passwd. This failed because I am not the owner of that file. Ive also tried to change the group owner of the file tmp, of which I am the owner, to the group man. However I am not a member of the group man so it has also failed.

chown and chgrp

The commands chown and chgrp are used to change the owner and group owner of a file. Format
chown [-R] owner files chgrp [-R] group files

The optional switch -R works in the same was as the -R switch for chmod. It modifies the command so that it descends any directories and performs the command on those sub-directories and files in those sub-directories. owner is either a numeric user identifier or a username. group is either a numeric group identifier or a group name. files is a list of files of which you wish to change the ownership. Some systems (Linux included) allow owner in the chown command to take the format owner.group. This allows you to change the owner and the group owner of a file with one command. Examples
chown david /home/david

Change the owner of the directory /home/david to david. This demonstrates one of the primary uses of the chown command. When a new account is created the root user creates a number of directories and files. Since root created them they are owned by root. In real life these files and directories should be owned by the new username.
chown -R root /

Change the owner of all files to root.

chown david.users /home/david Change the ownership of the file /home/david user david and the group users. chgrp users /home/david

so that it is owned by the

Change the group owner of the directory /home/david to the group users.

David Jones (20.01.00)

Page 124

85321, Systems Administration

Chapter 5: Processes and Files

Default permissions
When you create a new file it automatically receives a set of file permissions.
dinbig:~$ touch testing dinbig:~$ ls -l testing -rw-r--r-1 david users

0 Feb 10 17:36 testing

In this example the touch command has been used to create an empty file called testing. The touch command usually updates the last modified date of a file to the current time. However, if the file doesnt exist it creates an empty file of that same name. In this example the file testing has been given the default permissions rw-r-r--. Any file I create will receive the same default permissions.

umask
The built-in shell command umask is used specify and view what the default file permissions are. Executing the umask command without any arguments will cause it to display what the current default permissions are.
dinbig:~$ umask 022

By default the umask command uses the numeric format for permissions. It returns a number which specifies which permissions are turned off when a file is created. In the above example the user has the value 0 This means that by default no permissions are turned off for the user. the group and other have the value 2 This means that by default the write permission is turned off.

You will notice that the even though the execute permission is not turned off my default file doesnt have the execute permission turned on. I am not aware of the exact reason for this.
umask

versions

Since umask is a built-in shell command the operation of the umask command will depend on the shell you are using. This also means that youll have to look at the man page for your shell to find information about the umask command.
umask

for bash

The standard shell for Linux is bash. The version of umask for this shell supports symbolic permissions as well as numeric permissions. This allows you to perform the following.
dinbig:~$ umask -S u=rwx,g=r,o=r dinbig:~$ umask u=rw,g=rw,o=

David Jones (20.01.00)

Page 125

85321, Systems Administration

Chapter 5: Processes and Files

dinbig:~$ umask -S u=rw,g=rw,o=

Exercises
5.13.

Use the umask command so that the default permissions for new files are set to
rw------772

5.14.

How do you test that this is working correctly?

File permissions and directories

As shown in table 5.2 file permissions have a slightly different effect on directories than they do on files. The following example is designed to reinforce your understanding of the effect of file permissions on directories.

For example
Assume that I have an account on the same UNIX machine as you we belong to different groups I want to allow you to access the text for assignment one I want you to copy your finished assignments into my directory But I dont want you to see anything else in my directories

The following diagram represents part of my directory hierarchy including the file permissions for each directory.

Figure 5.4 Permissions and Directories

What happens if?

What happens if you try the following commands

David Jones (20.01.00)

Page 126

85321, Systems Administration

Chapter 5: Processes and Files

ls -l david

To perform an ls you must have read permission on the directory. In this case you dont. Only myself, as the owner of the file has read permission, so only I can obtain a listing of the files in my directory.
cat david/phone.book

You're trying to have a look at my phone book but you can't. You have permission to do things to files in my directory because you have execute permission on the directory david. However the permissions on the phone.book file mean that only I can read it. The same things occurs if you try the command cp david/phone.book ~/phone.book. To the file system you are trying to do the same thing, read the file phone.book.
ls david/85321 The permissions are set up so you can get a listing of the files in the david/85321 directory. Notice you have read permission on the 85321 directory. cat david/85321/assign.txt

Here you're trying to have a look at the assignment text. This will work. You have read permission on the file so you can read it. You have execute permission on the directories 85321 and david which means you can gain access to files and directories within those directories (if the permissions on the files let you).
cat david/85321/solutions/assign1.sol

Trying to steal a look at the solutions? Well you have the permissions on the file to do this. But you don't have the permissions on the directory solutions so this will fail. What would happen if I executed this command chmod o+r david/85321/solutions This would add read permission for you to the directory solutions. Can you read the assign1.sol file now? No you can't. To read the file or do anything with a file you must have execute permission on the directory it is in.
cp my.assign david/85321/assign.txt

What's this? Trying to replace my assignment with one of your own? Will this work? No because you don't have write permission for the file assign.txt.

Links
As mentioned earlier in this chapter it is possible for a single UNIX file to be known by more than one name. This is achieved using links. This section provides a brief overview of the two types of links supported by Linux, hard and soft. Links are useful for a number of reasons simplify file paths Rather than ask users to remember that a certain data file is stored in the file

David Jones (20.01.00)

Page 127

85321, Systems Administration

Chapter 5: Processes and Files

/usr/local/www/data/1999/T3/85349/assign1/solutions.txt you can create a link /85349/a1/solutions.txt which be used by

people. Support different locations Some versions of UNIX and UNIX commands expect files or directories to be in different locations. For example, some expect the executable program sendmail to be in /usr/lib/sendmail while others expect it to be in /usr/sbin/sendmail. Under Linux /usr/lib/sendmail is a hard link to /usr/sbin/sendmail.

The idea behind links is that you can use a different name to refer to the same file without having to duplicate the data blocks for the file.

Creating Links
Links, both hard and soft links, are created using the ln command. By default ln creates hard links. The ln command has two main formats ln target [link-name] Where target specifies the name of the file being linked to and link-name is the optional name of the link to create. Some examples

[david@faile tmp]$ ln /etc/passwd [david@faile tmp]$ ln /etc/passwd fred [david@faile tmp]$ ls -il * /etc/passwd 308647 -rw-r--r-3 root root /etc/passwd 308647 -rw-r--r-3 root root 308647 -rw-r--r-3 root root

681 Jan 681 Jan 681 Jan

3 08:37 3 08:37 fred 3 08:37 passwd

With the first ln command the link-name is not specified. In this instance the ln command creates a hard link with the same name as the target (passwd). In the second example the link-name is specified. The output of the ls command shows an important point about hard links (remember by default ln creates hard links). Since I am using the -i switch of the ls command we are shown the inodes for the three files. Each of these filenames (/etc/passwd, fred and passwd) all point to the same file and so they have the same inode. ln target-list directory In this format the ln command allows you to specify multiple targets followed by a directory name. The end result is that the directory will end up containing hard links to each of the targets. The names of the hard links will match that of the targets. For example,

[david@faile tmp]$ ln /etc/passwd /usr/share/magic /etc/group . [david@faile tmp]$ ls -il total 184 308651 -rw-r--r-2 root root 459 Oct 7 14:24 group 227274 -rw-r--r-2 root root 173852 Aug 24 08:06 magic 308647 -rw-r--r-2 root root 681 Jan 3 08:37 passwd [david@faile tmp]$ ls -il /etc/passwd /usr/share/magic /etc/group 308651 -rw-r--r-2 root root 459 Oct 7 14:24 /etc/group 308647 -rw-r--r-2 root root 681 Jan 3 08:37 /etc/passwd 227274 -rw-r--r-2 root root 173852 Aug 24 08:06 /usr/share/magic

David Jones (20.01.00)

Page 128

85321, Systems Administration

Chapter 5: Processes and Files

In this example there are three targets, /etc/passwd /usr/share/magic and /etc/group, and the directory specified is . (the full stop character which indicates the current directory). Again you can see the connection between the targets and the hard links via the similarity in the inodes. The ln command supports a number of command-line switches which can be used to modify its operation in a number of different ways. The options include -b, the backup switch If there is already a file with the same name as the link this switch will make a backup of the existing file. -s, the soft link switch This forces ln to create a soft link rather than a hard link.

The manual page for the ln command explains further. Some examples
[david@faile tmp]$ ln -b /etc/group passwd [david@faile tmp]$ ln -sb /etc/group magic [david@faile tmp]$ ls -il total 188 308651 -rw-r--r-3 root root 459 Oct 7 14:24 group 1105693 lrwxrwxrwx 1 david david 10 Jan 9 14:34 magic -> /etc/group 227274 -rw-r--r-2 root root 173852 Aug 24 08:06 magic~ 308651 -rw-r--r-3 root root 459 Oct 7 14:24 passwd 308647 -rw-r--r-2 root root 681 Jan 3 08:37 passwd~

The first thing to notice in the output of the ls command are the two new files magic~ and passwd~. These are the backup files which were created by using the -b switch of the ln command. The ln command creates backups by adding a ~ to the filename. The other difference to notice is magic -> /etc/group output for the magic file. This is how the ls command indicates a soft link. Also notice how the file type for the symbolic link is now l.

Hard and soft links, the differences

The differences between hard and soft links can be placed into two categories implementation, and operational.

The implementation details will be discussed in more detail in the File System chapter later in the text and involves how the operating system actually implements hard and soft links. One of the major implementation differences is that hard links and the file they are pointing to refer to the same inode where as soft links have their own inodes. The difference in inodes is the reason behind the operational differences between the two types of links. Standard file operations which use or modify data which is stored on the inode will behave differently with soft links than they do with hard links. Remember soft links use different inodes. So if you change any information on the inode of a soft link that same change will not be made on the inode of the file the soft link is pointing to. However, making any

David Jones (20.01.00)

Page 129

85321, Systems Administration

Chapter 5: Processes and Files

change to the inode information of a hard link also changes the inode information for the file the link is pointing to. Earlier in this chapter we introduced the information which is stored on the inode including file permissions, modified date, file size etc. Lets create some links and compare the results.
[david@faile tmp]$ ln -s ../rec/file.jpg file [david@faile tmp]$ ln ../rec/anna.jpg anna [david@faile tmp]$ ls -il file anna ../rec/anna.jpg ../rec/file.jpg 537951 -rw-rw-r-2 david david 5644 Dec 20 19:13 ../rec/anna.jpg 537955 -rwxrwxr-x 1 david david 17833 Dec 20 19:20 ../rec/file.jpg 537951 -rw-rw-r-2 david david 5644 Dec 20 19:13 anna 1105712 lrwxrwxrwx 1 david david 15 Jan 9 15:19 file -> ../rec/file.jpg

Notice how the information displayed about the hard link (anna) and the file it points to (../rec/anna.jpg) are exactly the same. This idicates a hard link. Notice how the information about the soft link (file) and the file it points to (../rec/file.jpg) are different?
[david@faile tmp]$ chmod a+x anna [david@faile tmp]$ chmod a-x file [david@faile tmp]$ ls -il file anna ../rec/anna.jpg ../rec/file.jpg 537951 -rwxrwxr-x 2 david david 5644 Dec 20 19:13 ../rec/anna.jpg 537955 -rw-rw-r-1 david david 17833 Dec 20 19:20 ../rec/file.jpg 537951 -rwxrwxr-x 2 david david 5644 Dec 20 19:13 anna 1105712 lrwxrwxrwx 1 david david 15 Jan 9 15:19 file -> ../rec/file.jpg

The chmod command changes the permissions associated with a file. From the above you can see that if you change the file permissions for a hard link that the permissions for both the hard link and the file it is pointing to change. However for the software link only the permissions for the file being pointed to change. Another important difference between hard and soft links is what happens when you start deleting files.
[david@faile tmp]$ rm ../rec/file.jpg ../rec/anna.jpg [david@faile tmp]$ ls -il file anna ../rec/anna.jpg ../rec/file.jpg ls: ../rec/anna.jpg: No such file or directory ls: ../rec/file.jpg: No such file or directory 537951 -rwxrwxr-x 1 david david 5644 Dec 20 19:13 anna 1105712 lrwxrwxrwx 1 david david 15 Jan 9 15:19 file -> ../rec/file.jpg [david@faile tmp]$ file anna file anna: JPEG image data, JFIF standard file: broken symbolic link to ../rec/file.jpg

When you remove a hard link or the file it points to it simply reduces the link count. The link count is the first number after the file permissions. Notice how the file anna has a link count of 1 in the above example where it had a 2 in earlier examples. The file anna still has a link to the data files contained in the image anna.jpg. On the other hand when you remove the file a soft link points to you are in trouble. The symbolic link does not know that the target has disappeared. The output of the file command shows this.
David Jones (20.01.00) Page 130

85321, Systems Administration

Chapter 5: Processes and Files

Searching the file hierarchy

A common task for a Systems Administrator is searching the UNIX file hierarchy for files which match certain criteria. Some common examples of what and why a Systems Administrator may wish to do this include searching for very large files finding where on the disk a particular file is deleting all the files owned by a particular user displaying the names of all files modified in the last two days.

Given the size of the UNIX file hierarchy and the number of files it contains this isnt a task that can be done by hand. This is where the find command becomes useful.

The find command

The find command is used to search through the directories of a file system looking for files that match a specific criteria. Once a file matching the criteria is found the find command can be told to perform a number of different tasks including running any UNIX command on the file.
find

command format

The format for the find command is

find [path-list] [expression] path-list is a list of directories in which the find command will search for files. The command will recursively descend through all sub-directories under these directories. The expression component is explained in the next section. Some examples
[root@faile tmp]# find /etc -type s

The path-list contains just /etc while the expression is -type s. In this case there are no files underneath the /etc directory which match the expression.
[root@faile tmp]# find /etc /usr/share -name magic -o -name passwd /etc/passwd /etc/pam.d/passwd /etc/uucp/passwd /usr/share/magic

In this example the path-list contains the directories /etc and /usr/share and the expression is -name magic -o -name passwd. This command finds four files under either the /etc or /usr/share directories which have the filename passwd or magic. Both the path and the expression are optional. If you run the find command without any parameters it uses a default path, the current directory, and a default expression, print the name of the file. The following is an example of what happens

David Jones (20.01.00)

Page 131

85321, Systems Administration

Chapter 5: Processes and Files

dinbig:~$ find . ./iAm ./iAm.c ./parameters ./numbers ./pass ./func ./func2 ./func3 ./pattern ./Adirectory ./Adirectory/oneFile

The default path is the current directory. In this example the find command has recursively searched through all the directories within the current directory. The default expression is -print. This is a find command that tells the find command to display the name of all the files it found. Since there was no test specified the find command matched all files. find expressions A find expression can contain the following components options, These modify the way in which the find command operates. tests, These decide whether or not the current file is the one you are looking for. actions, Specify what to do once a file has been selected by the tests. and operators. Used to group expressions together. options

find

Options are normally placed at the start of an expression. Table 5.6 summarises some of the find commands options. Option Effect
-daystart -depth -maxdepth number

for tests using time measure time from the beginning of today process the contents of a directory before the directory is a positive integer that specifies the maximum number of directories to descend
number number is a positive integer that specifies at which level to start applying tests

-mindepth number

-mount -xdev

dont cross over to other partitions dont cross over to other partitions
Table 5.6

find options

David Jones (20.01.00)

Page 132

85321, Systems Administration

Chapter 5: Processes and Files

For example The following are two examples of using finds options. Since I dont specify a path in which to start searching the default value, the current directory, is used.
dinbig:~$ find -mindepth 2 ./Adirectory/oneFile

In this example the mindepth option tells find to only find files or directories which are at least two directories below the starting point.
dinbig:~$ find -maxdepth 1 . ./iAm ./iAm.c ./parameters ./numbers ./pass ./func ./func2 ./func3 ./pattern ./Adirectory

This option restricts find to those files which are in the current directory.
find

tests

Tests are used to find particular files based on when the file was last accessed when the files status was last changed when the file was last modified the size of the file the files type the owner or group owner of the file the files name the files inode number the number and type of links the file has to it the files permissions

Table 5.7 summarises finds tests. A number of the tests take numeric values. For example, the number of days since a file was modified. For these situations the numeric value can be specified using one of the following formats (in the following n is a number)
+n

greater than n
-n

less than n

David Jones (20.01.00)

Page 133

85321, Systems Administration

Chapter 5: Processes and Files

equal to n For example Some examples of using tests are shown below. Note that in all these examples no command is used. Therefore the find command uses the default command which is to print the names of the files.
find . -user david

Find all the files under the current directory owned by the user david
find / -name \*.html

Find all the files one the entire file system that end in .html. Notice that the * must be quoted so that the shell doesnt interpret it (explained in more detail below). Instead we want the shell to pass the *.html to the find command and have it match filenames. are greater than 2500 kilobytes in size and have been in modified in the last seven days. The last example shows it is possible to combine multiple tests. It is also an example of using numeric values. The +2500 will match any value greater than 2500. The -7 will match any value less than 7. Shell special characters The shell is the program which implements the UNIX command line interface at which you use these commands. Before executing commands the shell looks for special characters. If it finds any it performs some special operations. In some cases, like the previous command, you dont want the shell to do this. So you quote the special characters. This process is explained in more detail in the following chapter.
find /home -size +2500k -mtime -7 Find all the files under the /home directory that

David Jones (20.01.00)

Page 134

85321, Systems Administration

Chapter 5: Processes and Files

Test
-amin n -anewer file -atime n -cmin n -cnewer file -ctime n -mmin n -mtime n -name pattern

Effect file last access n minutes ago the current file was access more recently than file file last accessed n days ago files status was changed n minutes ago the current files status was changed more recently than files files status was last changed n days ago files data was last modified n minutes ago the current files data was modified n days ago the name of the file matches pattern -iname is a case insensitive version of name -regex allows the use of REs to match filename the files UID or GID does not match a valid user or group the files permissions match mode (either symbolic or numeric) the file uses n units of space, b is blocks, c is bytes, k is kilobytes the file is of type c where c can be block device file, character device file, directory, named pipe, regular file, symbolic link, socket the files UID or GID matches n the file is owned by the user with name uname
Table 5.7

-nouser-nogroup -perm mode -size n[bck] -type c

-uid n -gid n -user uname

find tests find

actions

Once youve found the files you were looking for you want to do something with them. The find command provides a number of actions most of which allow you to either execute a command on the file, or display the name and other information about the file in a variety of formats

For the various find actions that display information about the file you are urged to examine the manual page for find Executing a command
find has two actions -exec and -ok.

that will execute a command on the files found. They are

The format to use them is as follows

David Jones (20.01.00)

Page 135

85321, Systems Administration

Chapter 5: Processes and Files

-exec command ; -ok command ;

is any UNIX command. The main difference between exec and ok is that ok will ask the user before executing the command. exec just does it.
command

For example Some examples of using the exec and ok actions include
find . -exec grep hello \{\} \;

Search all the files under the local directory for the word hello.
find / -name \*.bak -ok rm \{\} \; Find all files ending with .bak and ask the user if they wish

to delete those

files.
{}

and ;

The exec and ok actions of the find command make special use of {} and ; characters. All the characters here ( { and } and ; ) all have special meaning to the shell and as a result they must be quoted when used with the find command. The find command uses {} to refer to the filename that find has just tested. So in the last example rm \{\} will delete each file that the find tests match. The ; is used to indicate the end of the command to be executed.

Exercises
5.15.

Use find to print the names of every file on your file system that has nothing in it find where the file XF86Config is Write find commands to find all the files called core and display their full details find all the files called core which havent been accessed in the last week and delete them find all the files which are greater than 1Mb in size, were created in the last month and are not owned by the root user Write find commands to find all the files called core and display their full details find all the files called core which havent been accessed in the last week and delete them find all the files which are greater than 1Mb in size, were created in the last month and are not owned by the root user

5.16.

5.17.

David Jones (20.01.00)

Page 136

85321, Systems Administration

Chapter 5: Processes and Files

Performing commands on many files

Every UNIX command you execute requires a new process to be created. Creating a new process is a fairly heavyweight procedure for the operating system and can take quite some time. When you are performing a task it can save time if you minimise the number of new processes which are created. It is common for a Systems Administrator to want to perform some task which requires a large number of processes. Some uses of the find command offer a good example. For example Take the requirement to find all the HTML files on a Web site which contain the word expired. There are at least three different ways we can do this using the find command and the -exec switch, using the find command and back quotes , using the find command and the xargs command. More than one way to do something One of the characteristics of the UNIX operating system is that there is always more than one way to perform some task.
find

In the following well look at each of these.

and -exec

Well assume the files we are talking about in each of these examples are contained in the directory /usr/local/www
find /usr/local/www -name \*.html -exec grep -l expired \{\} \;

The -l switch of grep causes it to display the filename of any file in which it finds a match. So this command will list the names of all the files containing expired. While this works there is a slight problem, it is inefficient. These commands work as follows
find

searches through the directory structure,

every time it finds a file that matches the test (in this example that it has the extension html) it will run the appropriate command the operating system creates a new process for the command, once the command has executed for that file it dies and the operating system must clean up, now we restart at the top with find looking for the appropriate file

On any decent Web site it is possible that there will be tens and even hundreds of thousands of HTML files. For example, the 1999 website for 85321 contained 11,051 HTML files. This implies that this command will result in

David Jones (20.01.00)

Page 137

85321, Systems Administration

Chapter 5: Processes and Files

hundreds of thousands of processes being created. This can take quite some time.

find and back quotes

A solution to this is to find all the matching files first, and then pass them to a single grep command.
grep -l expired find /usr/local/www -name \*.html

In this example there are only two processes created. One for the find command and one for the grep. Back quotes Back quotes are an example of the shell special characters mentioned previously. When the shell sees characters it knows it must execute the command enclosed by the and then replace the command with the output of the command. In the above example the shell will execute the find command which is enclosed by the characters. It will then replace the
find /usr/local/www -name \*.html with the output of the command. Now the shell executes the grep command. Back quotes are explained in more detail in the next chapter.

To show the difference that this makes you can use the time command. time is used to record how long it takes for a command to finish (and a few other stats). The following is an example from which you can see the significant difference in time and resources used by reducing the number of processes.
beldin:~$ time grep -l expired find 85321/* -name index.html 0.04user 0.22system 0:02.86elapsed 9%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+0minor)pagefaults 0swaps beldin:~$ time find 85321/* -name index.html -exec grep -l expired \{\} \; 1.33user 1.90system 0:03.55elapsed 90%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (0major+0minor)pagefaults 0swaps

The time command can also report a great deal more information about a process and its interaction with the operating system. Especially if you use the verbose option (time v some_command)
find

and xargs

While in many cases the combination of find and back quotes will work perfectly, this method has one serious drawback as demonstrated in the following example.
beldin:~$ grep -l expired find 85321/* -name \* bash: /usr/bin/grep: Arg list too long

David Jones (20.01.00)

Page 138

85321, Systems Administration

Chapter 5: Processes and Files

The problem here is that a command line can only be so long. In the above example the find command found so many files that the names of these files exceeded the limit. This is where the xargs command enters the picture. Rather than pass the list of filenames as a parameter to the command, xargs allows the list of filenames to be passed as standard input (standard input is explained in more detail in a following chapter). This means we side-step the problem of exceeding the allowed maximum length of a command line.. Have a look at the man page for xargs for more information. Here is the example rewritten to use xargs
find /usr/local/www -name \* | xargs grep -l expired

There are now three processes created, find, xargs and grep. However it does avoid the problem of the argument list being too long.

Conclusion
UNIX is a multi-user operating system and as such must provide mechanisms to uniquely identify users and protect the resources of one user from other users. Under UNIX users are uniquely identified by a username and a user identifier (UID). The relationship between username and UID is specified in the /etc/passwd file. UNIX also provides the ability to collect users into groups. A user belongs to at least one group specified in the /etc/passwd file but can also belong to other groups specified in the /etc/group file. Each group is identified by both a group name and a group identifier (GID). The relationship between group name and GID is specified in the /etc/group file. All work performed on a UNIX computer is performed by processes. Each process has a real UID/GID pair and an effective UID/GID pair. The real UID/GID match the UID/GID of the user who started the process and are used for accounting purposes. The effective UID/GID are used for deciding the permissions of the process. While the effective UID/GID are normally the same as the real UID/GID it is possible using the setuid/setgid file permissions to change the effective UID/GID so that it matches the UID and GID of the file containing the process code. The UNIX file system uses a data structure called an inode to store information about a file including file type, file permissions, UID, GID, number of links, file size, date last modified and where the files data is stored on disk. A files name is stored in the directory which contains it. A files permissions can be represented using either symbolic or numeric modes. Valid operations on a file include read, write and execute. Users wishing to perform an operation on a file belong to one of three categories the user who owns the file, the group that owns the file and anyone (other) not in the first two categories. A files permissions can only be changed by the user who owns the file and are changed using the chmod command. The owner of a file can only be changed by the root user using the chown command. The group owner of a file can be

David Jones (20.01.00)

Page 139

85321, Systems Administration

Chapter 5: Processes and Files

changed by root user or by the owner of the file using the chgrp command. The files owner can only change the group to another group she belongs to. Links both hard and soft are mechanisms by which more than one filename can be used to refer to the same file.

Review Questions
5.1 For each of the following commands indicate whether they are built-in shell commands, "normal" UNIX commands or not valid commands. If they are "normal" UNIX commands indicate where the commands executable program is located. alias
!

history
"

rename
#

last
$

5.2 How would you find out what your UID, GID and the groups you currently belong to? 5.3 Assume that you are logged in with the username david and that your current directory contains the following files
bash# ls il total 2 103807 -rw-r--r-103808 -rwsr--r-103806 -rw-r--r-103807 -rw-r--r-2 1 1 2 david root david david users 0 Aug users 0 Aug users 2032 Aug users 0 Aug 25 25 22 25 13:24 14:11 11:42 13:24 agenda.doc meeting minutes.txt old_agenda

For each of the following commands indicate whether or not it will work,
% & '

if it works specify how the above directory listing will change, if it doesnt work why?

chmod 777 minutes.txt chmod u+w agenda.doc chmod o-x meeting chmod u+s minutes.txt ln -s meeting new_meeting chown root old_agenda

David Jones (20.01.00)

Page 140

85321, Systems Administration

Chapter 5: Processes and Files

5.4 Assume that the following files exist in the current directory.
bash$ ls -li total 1 32845 -rw-r--r-32845 -rw-r--r-32847 lrwxr-xr-x cq_uni_doc

2 jonesd 2 jonesd 1 jonesd

users users users

0 Apr 0 Apr 10 Apr

6 15:38 cq_uni_doc 6 15:38 cqu_union 6 15:38 osborne ->

For each of the following commands explain how the output of the command will change AFTER the command has been executed. Assume that that each command starts with the above information For example, after the command mv cq_uni_doc CQ.DOC the only change would be that entry for the file cq_uni_doc would change to
ls -li 32845 -rw-r--r-2 jonesd users 0 Apr 6 15:38 CQ.DOC

chmod a-x osborne chmod 770 cqu_union rm cqu_union rm cqu_uni_doc

The files cq_uni_doc and cqu_union both point to the same file using a hard link. Above I have stated that if you execute the command mv cq_uni_doc CQ.DOC the only thing that changes is the name of the file cq_uni_doc. Why doesnt the name of the file cqu_union change also?

David Jones (20.01.00)

Page 141

85321, Systems Administration

Chapter 6: The Shell

Chapter
The Shell

Introduction
You will hear many people complain that the UNIX operating system is hard to use. They are wrong. What they actually mean to say is that the UNIX command line interface is difficult to use. This is the interface that many people think is UNIX. In fact, this command line interface, provided by a program called a shell, is not the UNIX operating system and it is only one of the many different interfaces that you can use to perform tasks under UNIX. By this stage many of you will have used some of the graphical user interfaces provided by the X-Windows system. The shell interface is a powerful tool for a Systems Administrator and one that is often used. This chapter introduces you to the shell, its facilities and advantages. It is important to realise that the shell is just another UNIX command and that there are many different sorts of shell. The responsibilities of the shell include providing the command line interface
( ) 0 1 2 3 4 5 6

performing I/O redirection performing filename substitution performing variable substitution and providing an interpreted programming language

The aim of this chapter is to introduce you to the shell and the first four of the responsibilities listed above. The interpreted programming language provided by a shell is the topic of chapter 8.

Executing Commands
As mentioned previously the commands you use such as ls and cd are stored on a UNIX computer as executable files. How are these files executed? This is one of the major responsibilities of a shell. The command line interface at which you type commands is provided by the particular shell program you are using (under Linux you will usually be using a shell called bash). When you type a command at this interface and hit enter the shell performs the following steps wait for the user to enter a command perform a number of tasks if the command contains any special characters find the executable file for the command, if the file can't be found generate an error message fork off a child process that will execute the command,
Page 142

David Jones (20.01.00)

85321, Systems Administration

Chapter 6: The Shell

wait until the command is finished (the child process dies) and then return to the top of the list
7 8 9

Different shells
There are many different types of shells. Table 6.1 provides a list of some of the more popular UNIX shells. Under Linux most users will be using bash, the Bourne Again Shell. bash is an extension of the Bourne shell and uses the Bourne shell syntax. All of the examples in this text are written using the bash syntax. All shells fulfil the same basic responsibilities. The main differences between shells include the extra features provided Many shells provide command history, command line editing, command completion and other special features. the syntax Different shells use slightly different syntax for some commands. Program name
sh

Shell Bourne shell C shell Korn shell Bourne again shell

Description the original shell from AT&T, available on all UNIX machines shell developed as part of BSD UNIX AT&T improvement of the Bourne shell Shell distributed with Linux, version of Bourne shell that includes command line editing and other nice things

csh ksh bash

Table 6.1 Different UNIX shells

The C shell and its various versions have been popular in some fields. However, there are a number of problems with the C shell. The 85321 Website contains a pointer to a document entitled "C Shell Considered Harmful". If you really want to know why we use the Bourne shell syntax read this document.

Starting a shell
When you log onto a UNIX machine the UNIX login process automatically executes a shell for you. Which shell is executed is defined in the last field of your entry in the /etc/passwd file. The last field of every line of /etc/passwd specifies which program to execute when the user logs in. The program is usually a shell (but it doesnt have to be).

David Jones (20.01.00)

Page 143

85321, Systems Administration

Chapter 6: The Shell

Exercises
6.1.

What shell is started when you login?

The shell itself is just another executable program. This means you can choose to run another shell in the same way you would run any other command by simply typing in the name of the executable file. When you do the shell you are currently running will find the program and execute it. To exit a shell any of the following may work (depending on how your environment is set up). logout
@ A B

exit CTRL-D By default control D is the end of file (EOF) marker in UNIX. By pressing CTRL-D you are telling the shell that it has reached the end of the file and so it exits. In a later chapter which examines shell programming you will see why shells work with files.

For example The following is a simple example of starting other shells. Most different shells use a different command-line prompt.
bash$ sh $ csh % tcsh > exit % $ bash$

In the above my original login shell is bash. A number of different shells are then started up. Each new shell in this example changes the prompt (this doesnt always happen). After starting up the tcsh shell Ive then exited out of all the new shells and returned to the original bash.

Parsing the command line

The first task the shell performs when you enter a command is to parse the command line. This means the shell takes what you typed in and breaks it up into components and also changes the command-line if certain special characters exist. Special characters are used for a number of purposes and are used to modify the operation of the shell. Table 6.2 lists most of the special characters which the shell recognises and the meaning the shell places on these characters. In the following discussion the effect of this meaning and what the shell does with these special characters will be explained in more detail.

David Jones (20.01.00)

Page 144

85321, Systems Administration

Chapter 6: The Shell

Character(s)
white space

Meaning Any white space characters (tabs, spaces) are used to separate arguments multiple white space characters are ignored used to indicate the end of the commandline special quote characters that change the way the shell interprets special characters Used after a command, tells the shell to run the command in the background I/O redirection characters filename substitution characters indicate a shell variable used to separate multiple commands on the one line
Shell Table 6.2 special characters

newline character

" \

&

< >> << | * ? [ ] [^ $ ;

The Command Line

The following section examines, and attempts to explain, the special shell characters which influence the command line. This influence includes breaking the command line into arguments
C D E

allows more than one command to a line allows commands to be run in the background

Arguments
One of the first steps for the shell is to break the line of text entered by the user into arguments. This is usually the task of whitespace characters. What will the following command display?
echo hello there my friend

It wont display
hello there my friend

instead it will display

hello there my friend

When the shell examines the text of a command it divides it into the command and a list of arguments. A white space character separates the command and each argument. Any duplicate white space characters are ignored. The following diagram demonstrates.

David Jones (20.01.00)

Page 145

85321, Systems Administration

Chapter 6: The Shell

Shells,

Figure 6.1 white space and arguments

Eventually the shell will execute the command. The shell passes to the command a list of arguments. The command then proceeds to perform its function. In the case above the command the user entered was the echo command. The purpose of the echo command is to display each of its arguments onto the screen separated by a single space character. The important part here is that the echo command never sees all the extra space characters between hello and there. The shell removes this whilst it is performing its parsing of the command line.

One command to a line

The second shell special character in Table 6.2 is the newline character. The newline character tells the shell that the user has finished entering a command and that the shell should start parsing and then executing the command. The shell makes a number of assumptions about the command line a user has entered including there is only one command to each line
F G

the shell should not present the next command prompt until the command the user entered is finished executing.

This section examines how some of the shell special characters can be used to change these assumptions. Multiple commands to a line The ; character can be used to place multiple commands onto the one line.
ls ; cd /etc ; ls

The shell sees the ; characters and knows that this indicates the end of one command and the start of another.

Commands in the background

By default the shell will wait until the command it is running for the user has finished executing before presenting the next command line prompt. This
David Jones (20.01.00) Page 146

85321, Systems Administration

Chapter 6: The Shell

default operation can be changed by using the & character. The & character tells the shell that it should immediately present the next command line prompt and run the command in the background. This provides major benefits if the command you are executing is going to take a long time to complete. Running it in the background allows you to go on and perform other commands without having to wait for it to complete. However, you wont wish to use this all the time as some confusion between the output of the command running in the background and shell command prompt can occur. For example The sleep command usually takes on argument, a number. This number represents the number of seconds the sleep command should wait before finishing. Try the following commands on your system to see the difference the & character can make.
bash$ sleep 10 bash$ sleep 10 &

Filename substitution
In the great majority of situations you will want to use UNIX commands to manipulate files and directories in some way. To make it easier to manipulate large numbers of commands the UNIX shell recognises a number of characters which should be replaced by filenames. This process is called ether filename substitution or filename globbing. For example You have a directory which contains HTML files (an extension of .html), GIF files (an extension of .gif), JPEG files (an extension .jpg) and a range of other files. You wish to find out how big all the HTML files are. The hard way to do this is to use the ls l command and type in all the filenames. The simple method is to use the shell special character *, which represents any 0 or more characters in a file name
ls l *.html

In the above, the shell sees the * character and recognises it as a shell special character. The shell knows that it should replace *.html with any files that have filenames which match. That is, have 0 or more characters, followed by .html

David Jones (20.01.00)

Page 147

85321, Systems Administration

Chapter 6: The Shell

UNIX doesnt use extensions MS-DOS and Windows treat a file's extension as special. UNIX does not do this. Refer to the previous chapter and its discussion of magic numbers. Table 6.3 lists the other shell special characters which are used in filename substitution. Character
* ? [ ] [^ ]

What it matches
0 or more characters 1 character matches any one character between the brackets matches any one character NOT in the brackets
Table 6.3 substitution special characters

Filename

Some examples of filename substitution include

H I P Q R

cat * * will be replaced

by the names of all the files and directories in the current directory. The cat command will then display the contents of all those files.

ls a*bc a*bc matches ls a?bc a?bc matches

all filenames that start with a, end with bc and have any characters in between. all filenames that start with a, end with bc and have only ONE character in between. any filename that starts with either a i or c followed by any other three letters.

ls [ic]??? [ic]??? matches ls [^ic]???

Same as the previous command but instead of any file that starts with i or c match any file that DOESN'T start with i or c.

David Jones (20.01.00)

Page 148

85321, Systems Administration

Chapter 6: The Shell

Exercises
6.2.

Given the following files in your current directory:

$ ls feb86 jan12.89 jan19.89 jan26.89 jan5.89 jan85 jan86 jan87 jan88 mar88 memo1 memo10 memo2 memo2.sv

What would be the output from the following commands?

echo echo echo echo echo echo echo echo echo echo * *[^0-9] m[a-df-z]* [A-Z]* jan* *.* ????? *89 jan?? feb?? mar?? [fjm][ae][bnr]

Removing special meaning

There will be times when you wont want to use the shell special characters as shell special characters. For example, what happens if you really do want to display
hello there my friend

How do you do it? It's for circumstances like this that the shell provides shell special characters called quotes. The quote characters " \ tell the shell to ignore the meaning of any shell special character. To display the above you could use the command
echo hello there my friend

The first quote character tells the shell to ignore the meaning of any special character between it and the next . In this case it will ignore the meaning of the multiple space characters. So the echo command receives one argument instead of four separate arguments. The following diagram demonstrates.

David Jones (20.01.00)

Page 149

85321, Systems Administration

Chapter 6: The Shell

Shells,

Figure 6.2 commands and quotes

Table 6.4 lists each of the shell quote characters, their names and how the influence the shell. Character Name Action

single quote the shell will ignore all special characters contained within a pair of single quotes double quote the shell will ignore all special characters EXCEPT $ \ contained within a pair of double quotes backslash the shell ignores any special character immediately following a backslash

"

Table 6.4 Quote characters

Examples with quotes Try the following commands and observe what happens
echo Im David.

This causes an error because the quote character must be used as one of a pair. Since this line doesnt have a second character the shell continues to ignore all the shell special characters it sees, including the new line character which indicates the end of a command.
S T U

echo I\m David.

This is the correct implementation of what was attempted above. The \ quote character is used to remove the special meaning of the character so it is used as a normal character
echo * echo *

David Jones (20.01.00)

Page 150

85321, Systems Administration

Chapter 6: The Shell

Exercises
6.3.

6.4.

Input/output redirection
As the name suggests input/output (I/O) redirection is about changing the source of input or destination of output. UNIX I/O redirection is very similar (in part) to MS-DOS I/O redirection (guess who stole from who). I/O redirection, when combined with the UNIX philosophy of writing commands to perform one task, is one of the most important and useful combinations in UNIX.

How it works
All I/O on a UNIX system is achieved using files. This includes I/O to the screen and from a keyboard. Every process under UNIX will open a number of

David Jones (20.01.00)

V W X Y ` a b

echo \*

The previous three show two different approaches to removing the special meaning from a single character.
echo one two three four echo one two three four echo "one two three four" echo hello there \ my name is david Here the \ is used to ignore the special

meaning of the newline character at the end of the first line. This will only work if the newline character is immediately after the \ character. Remember, the \ character only removes the special meaning from the next character.

echo files = ; ls echo files = \; ls

Since the special meaning of the ; character is removed by the \ character means that the shell no longer assumes there are two commands on this line. This means the ls characters are treated simply as normal characters, not a command which must be executed.

Create files with the following names

stars*

-top
hello my friend "goodbye"

Now delete them. As was mentioned in the previous chapter the {} and ; used in the exec and ok actions of the find command must be quoted. The normal way of doing this is to use the \ character to remove the special meaning. Why doesnt the use of the single quote character work. e.g. why the following command doesnt work.
find . -name \*.bak -ok rm {} ;

Page 151

85321, Systems Administration

Chapter 6: The Shell

different files. To keep a track of the files it has, a process maintains a file descriptor for every file it is using.

File descriptors
A file descriptor is a small, non-negative integer. When a process reads/writes to/from a file it passes the kernel the file descriptor and asks it to perform the operation. The kernel knows which file the file descriptor refers to.

Standard file descriptors

Whenever the shell runs a new program (that is when it creates a new process) it automatically opens three file descriptors for the new process. These file descriptors are assigned the numbers 0, 1 and 2 (numbers from then on are used by file descriptors the process uses). The following table summarises their names, number and default destination. Name standard input (stdin) standard output (stdout) standard error (stderr) File descriptor 0 1 2 Default destination the keyboard the screen the screen

Table 6.5 Standard file descriptors

By default whenever a command asks for input it takes that input from standard input. Whenever it produces output it puts that output onto standard output and if the command generates errors then the error messages are placed onto standard error. For example, the ls command displays an error message when it cant find the file it was given.
[root@faile 85321]# ls /fred ls: /fred: No such file or directory

The "No such file or directory" message is sent to standard error.

Changing direction
By using the special characters in the table below it is possible to tell the shell to change the destination for standard input, output and error. For example
cat /etc/passwd > hello

tells the shell rather than send the contents of the /etc/passwd file to standard output, it should send it to a file called hello.

David Jones (20.01.00)

Page 152

85321, Systems Administration

Chapter 6: The Shell

Character(s)
Command < file Command > file

Result Take standard input from file Place output of command into file. Overwrite anything already in the file. Append the output of command into file. Take standard input for command from the following lines until a line that contains label by itself execute command and replace command with the output of the command pass the output of command1 to the input of
command2

Command >> file command << label

command

command1 | command2 command1 2> file

redirect standard error of command1 to file. The 2 can actually be replaced by any number which represents a file descriptor redirect output of command1 to a file_descriptor (the actual number for the file descriptor)

command1 >& file_descriptor

Table 6.6 I/O redirection constructs

Using standard I/O

Not all commands use standard input and standard output. For example the cd command doesnt take any input and doesnt produce any output. It simply takes the name of a directory as an argument and changes to that directory. It does however use standard error if it cant change into the directory. It doesnt make sense to redirect the I/O of some commands For example, the cp command doesnt produce any output. It may produce errors if it cannot copy the requested file but otherwise there is no output. So
cp /etc/passwd /tmp/passwd > output.dat

does not make sense.

Filters
On the other hand some commands will always take their input from standard input and put their output onto standard output. All of the filters discussed earlier in the textbook act this way. As an example lets take the cat command mentioned previously. If you execute the cat command without supplying it with any parameters it will take its input from standard input and place its output onto standard output.

David Jones (20.01.00)

Page 153

85321, Systems Administration

Chapter 6: The Shell

Try it. Execute the command cat with no arguments. Hit CTRL-D, on a line by itself,to signal the end of input. You should find that cat echoes back to the screen every line you type. Try the same experiment with the other filters mentioned earlier.

I/O redirection examples

c

ls > the.files Create a file the.files

that contains the list of files in the current

directory.
d

ls /fred 2> /dev/null

Send any error messages to the null device (throw it away).

cat the.files | more

Same effect as the command more the.files. Display the content of the file the.files one page at a time.
e f g h i p

ls /etc >> the.files

Add the list of files in from the /etc directory onto the end of the file
the.files. echo number of lines in the.files = wc -l the.files Execute the command wc -l the.files. Replace it with its output and then execute the echo command. Will display output similar to number of lines in the.files = 66 cat << finished > input

Ask the user to type in information until they enter a line with just finished on it. Then copy all the information entered by the user into the file called input
cd /etc > output.file

Create an empty file called output.file. The cd command generates no output so redirecting its output creates an empty file.
ls | cd

An error message. cd doesnt accept input so when the shell tries to send the output of the ls command to the cd command it doesnt work.
echo wc -l /etc/passwd Execute the wc command and pass

its output to the echo command as

arguments.

David Jones (20.01.00)

Page 154

85321, Systems Administration

Chapter 6: The Shell

Redirecting standard error

There will be times where you wish to either throw standard error away, join standard error and standard output, or just view standard error. This section provides examples of how this can be accomplished using I/O redirection. the file xx doesnt exist display an error message on standard error
$ ls xx /bin/ls: xx: No such file or directory

redirect standard output to the file errors, no change

$ ls xx > errors /bin/ls: xx: No such file or directory

redirect standard error to the file errors nothing on the screen file chap1.ps does exist so we get output but the errors still go to the file

$ ls xx 2> errors

$ ls chap1.ps xx 2> errors chap1.ps

try to send both stdout and stderr to the errors file, but stdout doesnt go

$ ls chap1.ps xx >& 2 2> errors chap1.ps

try a different order and it does work, why?

$ ls chap1.ps xx 2> errors >& 2

Evaluating from left to right

The shell evaluates arguments from left to right, that is it works with each argument starting with those from the left. This can influence how you might want to use the I/O redirection special characters. For example An example of why this is important is when you want to send both standard output and standard error of a command to the same file. Lets say we are attempting to view the attributes of the two files chap1.ps and xx. The idea is that the file xx does not exist so the ls command will generate an error when it cant find the file. Both the error and the file attributes of the chap1.ps file are meant to be sent to a file called errors. So we try to use
2>&1

This should redirect file descriptor 2 to standard output (refer back to Table 6.6). It should make standard error (file descriptor 2) go to the same place as standard output (file descriptor 1)
> output.and.errors

This sends standard output to the file output.and.errors (we hope). Lets try and find out.
David Jones (20.01.00) Page 155

85321, Systems Administration

Chapter 6: The Shell

$ ls -l chap1.ps xx 2>&1 > output.and.errors ls: xx: No such file or directory [david@faile tmp]$ cat output.and.errors -rw-rw-r-1 david david 0 Jan

9 16:23 chap1.ps

As you can see it doesnt work. The error message still appears on the screen and doesnt get sent to the output.and.errors file. Can you explain why? The reason it doesnt work is that the shell evaluates the arguments of this command from left to right. The order of evaluation goes like this ls The first argument tells the shell what command should be executed. -l The shell wont recognise any special characters in this argument so it will pass it on directly to the command. chap1.ps Again the shell wont see any shell special characters and so passes this argument directly onto the command. xx Same again. 2>&1 Now some action. The shell recognises some special characters here. It knows that >& are I/O redirection characters. These characters tell the shell that it should redirect standard error for this command to the same place as standard output. The current location for standard output is the terminal (the screen). So standard error is redirected to the terminal. No change from normal. > Again the shell will see a shell special character. In this case, the shell knows that standard output should be redirected to the location specified in the next argument. output.and.errors This is where the shell will send the standard error of the command, a file called output.and.errors.

The outcome of this is that standard output still goes to the terminal and standard error goes to the file output.and.errors. What we wanted is for both standard output and standard error to go to the file. The problem is the order in which the shell evaluated the arguments. The solution is to switch the I/O redirection shell characters.
[david@faile tmp]$ ls -l chap1.ps xx > output.and.errors 2>&1 [david@faile tmp]$ cat output.and.errors ls: xx: No such file or directory -rw-rw-r-1 david david 0 Jan 9 16:23 chap1.ps

David Jones (20.01.00)

Page 156

85321, Systems Administration

Chapter 6: The Shell

Changing the order means that standard output is redirected to the file output.and.errors FIRST and then standard error is redirected to where standard output is pointing (the same file).

Everything is a file
One of the features of the UNIX operating system is that almost everything can be treated as a file. This combined with I/O redirection allows you to achieve some powerful and interesting results. Youve already seen that by default stdin is the keyboard and stdout is the screen of your terminal. The UNIX operating system treats these devices as files (remember the shell sets up file descriptors for standard input/output). But which file is used?
tty

The tty command is used to display the filename of the terminal you are using.
$ tty /dev/ttyp1

In the above example my terminal is accessed through the file /dev/ttyp1. This means if I execute the following command
cat /etc/passwd > /dev/ttyp1

standard output will be redirected to /dev/ttyp1 which is where it wouldve gone anyway. Exercises
6.5.

What would the following command do? ls > tty

Device files
/dev/ttyp1 is an example of a device file. A device file is a interface to one of the kernels device drivers. A device driver is a part of the Linux kernel. It knows how to talk to a specific hardware device and presents a standard programming interface that is used by software. When you redirect I/O to/from a device file the information is passed through the device file, to the device driver and eventually to the hardware device or peripheral. In the previous example the contents of the /etc/passwd file were sent through the device file /dev/ttyp1, to a device driver. The device driver then displayed it on an appropriate device. /dev

All of the systems device files will be stored under the directory /dev. A standard Linux system is likely to have over 600 different device files. The following table summarises some of the device files.

David Jones (20.01.00)

Page 157

85321, Systems Administration

Chapter 6: The Shell

filename
/dev/hda

purpose The first IDE disk drive The first SCSI disk drive Sound card First floppy drive

filename
/dev/hda1

purpose the first partition on the first IDE disk drive the first partition on the first SCSI drive CD-ROM drive the second serial port

/dev/sda

/dev/sda1

/dev/audio /dev/fd0

/dev/cdrom /dev/ttyS1
files

Table 6.7 Example device

Redirecting I/O to device files

As youve seen it is possible to send output or obtain input from a device file. That particular example was fairly boring, heres another.
cat beam.au > /dev/audio

This one sends a sound file to the audio device. The result (if you have a sound card) is that the sound is played. When not to If you examine the file permissions of the device file /dev/hda1 youll find that only the root user and the group disk can write to that file. You should not be able to redirect I/O to/from that device file (unless you are the root user). If you could it would corrupt the information on the hard-drive. There are other device files that you should not experiment with. These other device file should also be protected with appropriate file permissions.
/dev/null /dev/null is the UNIX "garbage bin". Any output redirected to /dev/null is thrown away. Any input redirected from /dev/null is empty. /dev/null can be used to throw away output or create an empty file. cat /etc/passwd > /dev/null cat > newfile < /dev/null

The last command is one way of creating an empty file. Exercises

6.6.

Using I/O redirection how would you perform the following tasks - display the first field of the /etc/passwd file sorted in descending order - find the number of lines in the /etc/passwd file that contain the word
bash

David Jones (20.01.00)

Page 158

85321, Systems Administration

Chapter 6: The Shell

Shell variables
The shell provides a variable mechanism where you can store information for future use. Shell variables are used for two main purposes: shell programming and environment control. This section provides an introduction to shell variables and their use in environment control. A later chapter discusses shell programming in more detail.

Environment control
Whenever you run a shell it creates an environment. This environment includes pre-defined shell variables used to store special values including the format of the prompt the shell will present to you
q r s t u v w

your current path your home directory the type of terminal you are using and a great deal more.

Any shell variable you create will be stored within this environment. A later section in this chapter goes into more detail about environment control.

The set command

The set command can be used to view you shells environment. By executing the set command without any parameters it will display all the shell variables currently within your shells environment.

Using shell variables

There are two main operations performed with shell variables assign a variable a value use a variables value

Assigning a value
Assigning value to a shell variable is much the same as in any programming language variable_name=value.
my_variable=hello theNum=5 myName="David Jones"

A shell variable can be assigned just about any value, though there are a few guidelines to keep in mind. A space is a shell special character. If you want your shell variable to contain a space you must tell the shell to ignore the spaces special meaning. In the above example Ive used the double quotes. For the same reason there should never be any spaces around the = symbol.
David Jones (20.01.00) Page 159

85321, Systems Administration

Chapter 6: The Shell

Accessing a variables value

To access a shell variables value we use the $ symbol. The $ is a shell special character that indicates to the shell that it should replace a variable with its value. For example
dinbig$ myName="David Jones" dinbig$ echo My name is $myName My name is David Jones dinbig$ command=ls dinbig$ $command Mail ethics.txt papers dinbig$ echo A$empty: A:

Uninitialised variables
The last command in the above example demonstrates what the value of a variable is when you havent initialised it. The last command tries to access the value for the variable empty. But because the variable empty has never been initialised it is totally empty. Notice that the result of the command has nothing between the A and the :.

Resetting a variable
It is possible to reset the value of a variable as follows
myName=

This is totally different from trying this

myName=

This example sets the value of myName to a space character NOT nothing.

The readonly command

As you might assume the readonly command is used to make a shell variable readonly. Once you execute a command like
readonly my_variable

The shell variable my_variable can no longer be modified. To get a list of the shell variables that are currently set to read only you run the readonly command without any parameters.

The unset command

Previously youve been shown that to reset a shell variable to nothing as follows
variable=

David Jones (20.01.00)

Page 160

85321, Systems Administration

Chapter 6: The Shell

But what happens if you want to remove a shell variable from the current environment? This is where the unset command comes in. The command
unset variable

Will remove a variable completely from the current environment. There are some restrictions on the unset command. You cannot use unset on a read only variable or on the pre-defined variables IFS, PATH, PS1, PS2

Arithmetic
UNIX shells do not support any notion of numeric data types such as integer or real. All shell variables are strings. How then do you perform arithmetic with shell variables? One attempt might be
dinbig:~$ count=1 dinbig:~$ Rcount=$count+1

But it wont work. Think about what happens in the second line. The shell sees
$count and replaces it with the value of that variable so we get the command count=1+1. Since the shell has no notion of an integer data type the variable count now takes on the value 1+1 (just a string of characters).

The expr command

The UNIX command expr is used to evaluate expressions. In particular it can be used to evaluate integer expressions. For example
dinbig:~$ expr 5 + 6 11 dinbig:~$ expr 10 / 5 2 dinbig:~$ expr 5 \* 10 50 dinbig:~$ expr 5 + 6 * 10 expr: syntax error dinbig:~$ expr 5 + 6 \* 10 65

Note that the shell special character * has to be quoted. If it isnt the shell will replace it with the list of all the files in the current directory which results in expr generating a syntax error. Using expr By combining the expr command with the grave character we have a mechanism for performing arithmetic on shell variables. For example
count=1 count=expr $count + 1 expr

restrictions

The expr command only works with integer arithmetic. If you need to perform floating point arithmetic have a look at the bc and awk commands.

David Jones (20.01.00)

Page 161

85321, Systems Administration

Chapter 6: The Shell

The expr command accepts a list of parameters and then attempts to evaluate the expression they form. As with all UNIX commands the parameters for the expr command must be separated by spaces. If you dont expr interprets the input as a sequence of characters.
dinbig:~$ expr 5+6 5+6 dinbig:~$ expr 5+6 \* 10 expr: non-numeric argument

Alternatives to expr for arithmetic

The expr command is the traditional approach for perform arithmetic but it is by no means the best and has at least two major draw backs including it doesnt handle decimal points If you want to add 5.5 and 6.5 you cant do it with expr. One solution to this is the bc command

[david@faile tmp]$ echo 5.5 + 5 | bc 10.5

every use requires the creation of a new process Chapter 5 includes a discussion of why this can be a problem and cause shell scripts to be very slow. An alternative to this is to use the arithemetic capabilities provided by many of the modern shells including bash. This is what is used in the add2 script mentioned in the previous chapter.

[david@faile tmp]$ echo $[ 5 + 5] 10

Valid variable names

Most programming languages have rules that restrict the format of variable names. For the Bourne shell, variable names must start with either a letter or an underscore character, followed by zero or more letters, numbers or underscores

{}
In some cases you will wish to use the value of a shell variable as part of a larger word. Curly braces { } are used to separate the variable name from the rest of the word. For example You want to copy the file /etc/passwd into the directory /home/david. The following shell variables have been defined.
directory=/etc/ home=/home/david

A first attempt might be

David Jones (20.01.00)

Page 162

85321, Systems Administration

Chapter 6: The Shell

cp $directorypasswd $home

This wont work because the shell is looking for the shell variable called (there isnt one) instead of the variable directory. The correct solution would be to surround the variable name directory with curly braces. This indicates to the shell where the variable stops.
directorypasswd cp ${directory}passwd $home

Environment control
Whenever you run a shell it creates an environment in which it runs. This environment specifies various things about how the shell looks, feels and operates. To achieve this the shell uses a number of pre-defined shell variables. Table 6.8 summarises these special shell variables. Variable name Purpose
HOME SHELL

your home directory the executable program for the shell you are using your user id your username the type of terminal you are using your X-Windows display your executable path
Table 6.8 Environment variables

UID USER TERM DISPLAY PATH

PS1

and PS2

The shell variables PS1 and PS2 are used to store the value of your command prompt. Changing the values of PS1 and PS2 will change what your command prompt looks like.
dinbig:~$ echo :$PS1: and :$PS2: :\h:\w\$ : and :> :

is the secondary command prompt. It is used when a single command is spread over multiple lines. You can change the values of PS1 and PS2 just like you can any other shell variable.
PS2 bash

extensions

Youll notice that the value of PS1 above is \h:\w\$ but my command prompt looks like dinbig:~$. This is because the bash shell provides a number of extra facilities. One of those facilities is that it allows the command prompt to contain the hostname \h(the name of my machine) and the current working directory \w.

David Jones (20.01.00)

Page 163

85321, Systems Administration

Chapter 6: The Shell

With older shells it was not possible to get the command prompt to display the current working directory. Exercises
6.7.

Many first time users of older shells attempt to get the command prompt to contain the current directory by trying this PS1=pwd The pwd command displays the current working directory. Explain why this will not work. (HINT: When is the pwd command executed?)

Variables and sub-shells

Every time you start a new shell, the new shell will create a new environment separate from its parents environment. The new shell will not be able to access or modify the environment of its parent shell.

For example
Heres a simple example.
dinbig:~$ myName=david dinbig:~$ echo $myName david dinbig:~$ bash dinbig:~$ echo my name is $myName my name is dinbig:~$ exit

create a shell variable use it start a new shell try to use the parent shells variable exit from the new shell and return to the parent use the variable again

dinbig:~$ echo $myName david

As you can see a new shell cannot access or modify the shell variables of its parent shells.
export

There are times when you may wish a child or sub-shell to know about a shell variable from the parent shell. For this purpose you use the export command. For example,
dinbig:~$ myName=David Jones dinbig:~$ bash dinbig:~$ echo my name is $myName my name is dinbig:~$ logout dinbig:~$ export myName dinbig:~$ bash dinbig:~$ echo my name is $myName my name is david dinbig:~$ exit

David Jones (20.01.00)

Page 164

85321, Systems Administration

Chapter 6: The Shell

Local variables
When you export a variable to a child shell the child shell creates a local copy of the variable. Any modification to this local variable cannot be seen by the parent process. There is no way in which a child shell can modify a shell variable of a parent process. The export command only passes shell variables to child shells. It cannot be used to pass a shell variable from a child shell back to the parent. For example
dinbig:~$ echo my name is $myName my name is david dinbig:~$ export myName dinbig:~$ bash dinbig:~$ myName=fred # child shell modifies variable dinbig:~$ exit dinbig:~$ echo my name is $myName my name is david # there is no change in the parent

Advanced variable substitution

The shell provides a number of additional more complex constructs associated with variable substitution. The following table summarises them. Construct Purpose
${variable:-value}

replace this construct with the variables value if it has one, if it doesnt, use value but dont make variable equal to value same as the above but if variable has no value assign it value replace the construct with the value of the variable if it has one, if it doesnt then display message onto stderr if message is null then display prog: variable: parameter null or not set on stderr if variable has a value replace it with value otherwise do nothing
Table 6.9 Advanced variable substitution

${variable:=value}

${variable:?message}

${variable:+value}

For example
dinbig:~$ myName= dinbig:~$ echo my name my name is dinbig:~$ echo my name my name is NO NAME dinbig:~$ echo my name my name is dinbig:~$ echo my name my name is NO NAME dinbig:~$ echo my name
David Jones (20.01.00)

is $myName is ${myName:-"NO NAME"} is $myName is ${myName:="NO NAME"} is $myName

Page 165

85321, Systems Administration

Chapter 6: The Shell

my name is NO NAME dinbig:~$ herName= dinbig:~$ echo her name is ${herName:?"she hasnt got a name"} bash: herName: she hasnt got a name dinbig:~$ echo her name is ${herName:?} bash: herName: parameter null or not set

Evaluation order
In this chapter weve looked at the steps the shell performs between getting the users input and executing the command. The steps include I/O redirection Where the shell changes the direction in which I/O is being sent. variable substitution The shell replaces shell variables with the corresponding values. filename substitution This is where the shell replaces globbing characters with matching filenames.

An important question is in what order does the shell perform these steps?

Why order is important

Look at the following example
dinbig:~$ dinbig:~$ | dinbig:~$ dinbig:~$ Mail News pipe=\| echo $pipe star=\* echo $star README VMSpec.ps.bak acm.bhx acm2.dot

In the case of the echo $start command the shell has seen $star and replaced it with its value *. The shell sees the * and replaces it with the list of the files in the current directory. In the case of the echo $pipe command the shell sees $pipe and replaces it with its value |. It then displays | onto the screen. Why didnt it treat the | as a special character? If it had then echo | wouldve produced something like the following.
[david@faile tmp]$ echo | >

The >, produced by the shell not typed in by the user, indicates that the shell is still waiting for input. The shell is still expecting another command name. The reason this isnt produced in the previous example is related to the order in which the shell performs its analysis of shell special variables.

The order
The order in which the shell performs the steps is I/O redirection

David Jones (20.01.00)

Page 166

85321, Systems Administration

Chapter 6: The Shell

variable substitution filename substitution

For the command

echo $PIPE

the shell performs the following steps check for any I/O redirection characters, there arent any, the command line is currently echo $PIPE check for variables, there is one $PIPE, replace it with its value, the command line is now echo | check for any wildcards, there arent any

So it now executes the command echo |. If you do the same walk through for the echo $star command you should see how its output is achieved.

The eval command

What happens if I want to execute the following command

ls $pipe more

using the shell variable pipe from the example above? The intention is that the pipe shell variable should be replaced by its value | and that the | be used to redirect the output of the ls command to the more command. Due to the order in which the shell performs its evaluation this wont work.

Doing it twice
The eval command is used to evaluate the command line twice. eval is a built-in shell command. Take the following command (using the pipe shell variable from above)
eval ls $pipe more

The shell sees the $pipe and replaces it with its value, |. It then executes the eval command. The eval command repeats the shells analysis of its arguments. In this case it will see the | and perform necessary I/O redirection while running the commands.

Conclusion
The UNIX command line interface is provided by programs called shells. A shells responsibilities include providing the command line interface performing I/O redirection

David Jones (20.01.00)

Page 167

85321, Systems Administration

Chapter 6: The Shell

performing filename substitution performing variable substitution and providing an interpreted programming language

A shell recognises a number of characters as having special meaning. Whenever it sees these special characters it performs a number of tasks that replace the special characters. When a shell is executed it creates an environment in which to run. This environment consists of all the shell variables created including a number of pre-defined shell variables that control its operation and appearance.

Review Questions
6.1 What is the effect of the following command sequences?
ls | wc -l rm ??? who | wc -l mv progs/* /usr/steve/backup ls *.c | wc -l rm *.o who | sort cd ; pwd cp memo1 .. ls -l | sort +4n

6.2

What is the output of the following commands? Are there any problems? How would you fix it?
echo this is a star * echo ain\\\\t you my friend echo "** hello **" echo "the output of the ls command is ls" echo the output of the pwd command is pwd

6.3

Which of the following are valid shell variable names?

XxXxXxXx _ 12345

David Jones (20.01.00)

Page 168

85321, Systems Administration

Chapter 6: The Shell

HOMEDIR file.name _date file_name x0-9 file1 Slimit

6.4

Suppose your HOME directory is /usr/steve and that you have sub-directory as shown in figure 6.3. Assuming you just logged onto the system and executed the following commands:
docs=/usr/steve/documents let=$docs/letters prop=$docs/proposals

write commands to do the following using these variables

List the contents of the documents directory Copy all files from the letters directory to the proposals directory Move all files with names that contain a capital letter from the letters directory to the current directory. Count the number of files in the memos directory.
ls $let/.. cat $prop/sys.A >> $let/no.JSK echo $let/* cp $let/no.JSK $prop cd $prop files_in_prop=echo $prop* cat echo $let\*

What would be the effect of the following commands?

David Jones (20.01.00)

Page 169

85321, Systems Administration

Chapter 6: The Shell

Figure 6.3 Review Question 6.4

David Jones (20.01.00)

Page 170

85321, Systems Administration

Chapter 7: Text Manipulation

Chapter
Text Manipulation

Introduction
Many of the tasks a Systems Administrator will perform involve the manipulation of textual information. Some examples include manipulating system log files to generate reports and modifying shell programs. Manipulating textual information is something which UNIX is quite good at and provides a number of tools which make tasks like this quite simple, once you understand how to use the tools. The aim of this chapter is to provide you with an understanding of these tools By the end of this chapter you should be familiar with using regular expressions, able to use regular expressions and ex commands to perform powerful text manipulation tasks. Online lecture 7 on the 85321 Website/CD-ROM provides some alternative discussion of the topics covered in this chapter. It may be beneficial to follow that lecture in conjunction with reading this chapter.

Regular expressions
Regular expressions provide a powerful method for matching patterns of characters. Regular expressions (REs) are understood by a number of commands including ed ex sed awk grep egrep, expr and even vi. Some examples of regular expressions look like include
david

Will match any occurrence of the word david

[Dd]avid

Will match either david or David

.avid

Will match any letter (.) followed by avid

^david$

Will match any line that contains only david

d*avid

Will match avid, david, ddavid dddavid and any other word with repeated ds followed by avid
^[^abcef]avid$

Will match any line with only five characters on the line, where the last

David Jones (20.01.00)

Page 171

85321, Systems Administration

Chapter 7: Text Manipulation

four characters must be avid and the first character can be any character except abcef. Each regular expression is a pattern, it matches a collection of characters. That means by itself the regular expression can do nothing. It has to be combined with some UNIX commands which understand regular expressions. The simplest example of how regular expressions are used by commands is the grep command. The grep command was introduced in a previous chapter and is used to search through a file and find lines that contain particular patterns of characters. Once it finds such a line, by default, the grep command will display that line onto standard output. In that previous chapter you were told that grep stood for global regular expression pattern match. Hopefully you now have some idea of where the regular expression part comes in. This means that the patterns that grep searches for are regular expressions. The following are some example command lines making use of the grep command and regular expressions
grep unix tmp.doc

find any lines contain unix

grep [Uu]nix tmp.doc

find any lines containing either unix or Unix. Notice that the regular expression must be quoted. This is to prevent the shell from treating the [] as shell special characters and performing file name substitution.
grep [^aeiouAEIOU]* tmp.doc

Match any number of characters that do not contain a vowel.

grep ^abc$ tmp.doc

Match any line that contains only abc.

grep hel. tmp.doc Match hel followed by any other character.

Other UNIX commands which use regular expressions include sed, ex and vi. Which are editors (different types of editors) which allow the use of regular expressions to search and to search and replace patterns of characters. Much of the power of the Perl script language and the awk command can also be traced back to regular expressions. You will also find that the use of regular expressions on other platforms (i.e. Microsoft) is increasing as the benefits of REs become apparent.

REs versus filename substitution

It is important that you realise that regular expressions are different from filename substitution. If you look in the previous examples using grep you will see that the regular expressions are sometimes quoted. One example of this is the command
grep [^aeiouAEIOU]* tmp.doc

David Jones (20.01.00)

Page 172

85321, Systems Administration

Chapter 7: Text Manipulation

Remember that [^] and * are all shell special characters. If the quote characters () were not there the shell would perform filename substitution and replace these special characters with matching filenames. For example, if I execute the above command without the quote characters in one of the directories on my Linux machine the following happens
[david@faile tmp]$ grep [^aeiouAEIOU]* tmp.doc tmp.doc:chap1.ps this is the line to match

The output here indicates that grep found one line in the file tmp.doc which contained the regular expression pattern it wanted and it has displayed that line. However this output is wrong. Remember, before the command is executed the shell will look for and modify any shell special characters it can find. In this command line the shell will find the regular expression because it contains special characters. It replaces the [^aeiouAEIOU]* with all the files in the current directory which dont start with a vowel (aeiouAEIOU). The following sequence shows what is going on. First the ls command is used to find out what files are in the current directory. The echo command is then used to discover which filenames will be matched by the regular expression. You will notice how the file anna is not selected (it starts with an a). The grep command then shows how when you replace the attempted regular expression with what the shell will do you get the same output as the grep command above with the regular expression.
[david@faile tmp]$ ls anna chap1.ps magic tmp tmp.doc [david@faile tmp]$ echo [^aeiouAEIOU]* chap1.ps magic tmp tmp.doc [david@faile tmp]$ grep chap1.ps magic tmp tmp.doc tmp.doc:chap1.ps this is the line to match

In this example command we do not want this to happen. We want the shell to ignore these special characters and pass them to the grep command. The grep command understands regular expressions and will treat them as such. The output of the proper command on my system is
[david@faile tmp]$ grep [^aeiouAEIOU]* tmp.doc This is atest chap1.ps this is the line to match

Regular expressions have nothing to do with filename substitution, they are in fact completely different. Table 7.1 highlights the differences between regular expressions and filename substitution. Filename substitution Regular expressions

Performed by the shell Performed by individual commands used to match filenames Used to match patterns of characters in data files
Regular Table 7.1 expressions versus filename substitution

David Jones (20.01.00)

Page 173

85321, Systems Administration

Chapter 7: Text Manipulation

How they work

Regular expressions use a number of special characters to match patterns of characters. Table 7.2 outlines these special characters and the patterns they match. Character
c

Matches if c is any character other than \ [ . * ^ ] $ then it will match a single occurrence of that character remove the special meaning from the following character any one character the start of a line the end of a line 0 or more matches of the previous RE any one character in chars a list of characters any one character NOT in chars a list of characters
Regular Table 7.2 expression characters

. ^ $ * [chars]

[^chars]

Exercises
7.4.

What will the following simple regular expressions match? fred [^D]aily ..^end$ he..o he\.\.o \$fred $fred

Repetition
There are times when you wish to repeat a previous regular expression. For example, I want to match 40 letter as. One approach would be to write
aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa

Another would be to use one of the repetition characters from Table 7.3.
a\{40,40\}

David Jones (20.01.00)

Page 174

85321, Systems Administration

Chapter 7: Text Manipulation

Construct
+

Purpose match one or more occurrences of the previous RE match zero or one occurrences of the previous RE match exactly n occurrences of the previous RE match at least n occurrences of the previous RE match between n and m occurrences of the previous RE
Table 7.3 expression repetition characters

\{n\}

\{n,\}

\{n, m\}

Regular

Each of the repetition characters in the previous table will repeat the previous regular expression. For example,
d+

Match one or more ds

fred?

Match fre followed by 0 or more ds. NOT 0 or more repetitions of fred.

.\{5,\}

Does not match 5 or more repeats of the same character (e.g. aaaaa). Instead it matches at least 5 or more repeats of any single character. This last example is an important one. The repetition characters match the previous regular expression and NOT what the regular expression matches. The following commands show the distinction
[david@faile tmp]$ cat pattern aaaaaaaaaaa david dawn [david@faile tmp]$ grep .\{5,\} pattern aaaaaaaaaaa david

First step is to show the contents of the file pattern, three lines of text, one with a row of as, another with the name david and another with the name dawn. If the regular expression .\{5,\} is meant to match at least 5 occurrences of the same character it should only match the line with all as. However, as you can see it also matches the line containing david. The reason for this is that .\{5,\} will match any line with at least 5 single characters. So it does mention the line with the name david but doesnt match the line with the name dawn. That last line isnt matched because it only contains 4 characters.

David Jones (20.01.00)

Page 175

85321, Systems Administration

Chapter 7: Text Manipulation

Concatenation and Alternation

It is quite common to concatenate regular expressions one after the other. In this situation any string which the regular expression matches will match the entire regular expression. Alternation, choosing between two or regular expressions is done using the | character. For example,
x

egrep (a|b) pattern

Match any line that contains either an a or a b

Different commands, different REs

Regular expressions are one area in which the heterogeneous nature of UNIX becomes apparent. Different programs on different platforms recognise different subsets of regular expressions. You need to refer to the manual page of the various commands to find out which features it supports. On Linux you can also check the regex(7) manual page for more details about the POSIX 1003.2 regular expressions supported by most of the GNU commands used by Linux. One example of the difference, using the pattern file used above, follows
[david@faile tmp]$ grep .\{2,\} pattern aaaaaaaaaaa david [david@faile tmp]$ egrep .\{2,\} pattern

This demonstrates how the grep and egrep commands on Linux use slightly different versions of regular expressions.

Exercises
7.5.

Write grep commands that use REs to carry out the following. 1. Find any line starting with j in the file /etc/passwd (equivalent to asking to find any username that starts with j). 2. Find any user that has a username that starts with j and uses bash as their login shell (if they use bash their entry in /etc/passwd will end with the full path for the bash program). 3. Find any user that belongs to a group with a group ID between 0 and 99 (group id is the fourth field on each line in /etc/passwd).

Tagging
Tagging is an extension to regular expressions which allows you to recognise a particular pattern and store it away for future use. For example, consider the regular expression
da\(vid\)

David Jones (20.01.00)

Page 176

85321, Systems Administration

Chapter 7: Text Manipulation

The portion of the RE surrounded by the \( and \) is being tagged. Any pattern of characters that matches the tagged RE, in this case vid, will be stored in a register. The commands that support tagging provide a number of registers in which character patterns can be stored. It is possible to use the contents of a register in a RE. For example,
\(abc\)\1\1

The first part of this RE defines the pattern that will be tagged and placed into the first register (remember this pattern can be any regular expression). In this case the first register will contain abc. The 2 following \1 will be replaced by the contents of register number 1. So this particular example will match abcabcabc. The \ characters must be used to remove the other meaning which the brackets and numbers have in a regular expression.

For example
Some example REs using tagging include
\(david\)\1

This RE will match daviddavid. It first matches david and stores it into the first register (\(david\)). It then matches the contents of the first register (\1).
\(.\)oo\1

Will match words such as noon, moom.

pattern.

For the remaining RE examples and exercises Ill be referring to a file called The following is the contents of pattern.

a hellohello goodbye friend how hello there how are you how are you ab bb aaa lll Parameters param

Exercises
7.6.

What will the following commands do

grep \(a\)\1 pattern grep \(.*\)\1 pattern grep \( .*\)\1 pattern

ex, ed, sed

and vi

So far you've been introduced to what regular expressions do and how they work. In this section you will be introduced to some of the commands which allow you to use regular expressions to achieve some quite powerful results.
David Jones (20.01.00) Page 177

85321, Systems Administration

Chapter 7: Text Manipulation

In the days of yore UNIX did not have full screen editors. Instead the users of the day used the line editor ed. ed was the first UNIX editor and its impact can be seen in commands such as sed, awk, grep and a collection of editors including ex and vi. vi was written by Bill Joy while he was a graduate student at the University of California at Berkeley (a University responsible for many UNIX innovations). Bill went on to do other things including being involved in the creation of Sun Microsystems. vi is actually a full-screen version of ex. Whenever you use :wq to save and quit out of vi you are using a ex command.

So???
All very exciting stuff but what does it mean to you a trainee Systems Administrator? It actually has at least three major impacts by using vi you can become familiar with the ed commands commands allow you to use regular expressions to manipulate and modify text
ed

those same ed commands, with regular expressions, can be used with sed to perform all these tasks non-interactively (this means they can be automated).

Why use ed?

Why would anyone ever want to use a line editor like ed? Well in some instances the Systems Administrator doesnt have a choice. There are circumstances where you will not be able to use a full screen editor like vi. In these situations a line editor like ed or ex will be your only option. One example of this is when you boot a Linux machine with installation boot and root disks. A few years ago these disks usually didnt have space for a full screen editor but they did have ed.
ed ed

commands

is a line editor that recognises a number of commands that can manipulate text. Both vi and sed recognise these same commands. In vi whenever you use the : command you are using ed commands. ed commands use the following format.
[ address [, address]] command [parameters]

(you should be aware that anything between [] is optional) This means that every ed command consists of 0 or more addresses that specify which lines the command should be performed upon, a single character command, and an optional parameter (depending on the command)
Page 178

David Jones (20.01.00)

85321, Systems Administration

Chapter 7: Text Manipulation

For example Some example ed commands include

1,$s/old/new/g The address is 1,$

which specifies all lines. The command is the substitute command. With the following text forming the parameters to the command. This particular command will substitute all occurrences of the work old with the word new for all lines within the current file.
4d3

The address is line 4. The command is delete. The parameter 3 specifies how many lines to delete. This command will delete 3 lines starting from line 4.
d

Same command, delete but no address or parameters. The default address is the current line and the default number of lines to delete is one. So this command deletes the current line.
1,10w/tmp/hello

The address is from line 1 to line 10. The command is write to file. This command will write lines 1 to 10 into the file /tmp/hello The current line The ed family of editors keep track of the current line. By default any ed command is performed on the current line. Using the address mechanism it is possible to specify another line or a range of lines on which the command should be performed. Table 7.4 summarises the possible formats for ed addresses.

David Jones (20.01.00)

Page 179

85321, Systems Administration

Chapter 7: Text Manipulation

Address
. $ 7 a /RE/

Purpose the current line the last line line 7, any number matches that line number the line that has been marked as a the next line matching the RE moving forward from the current line the next line matching the RE moving backward from the current line the line that is n lines after the line specified by address the line that is n lines before the line specified by
address

?RE?

Address+n Address-n Address1, address2 ,

a range of lines from address1 to address2 the same as 1,$, i.e. the entire file from line 1 to the last line ($) the same as .,$, i.e. from the current line (.) to the last line ($)
Table 7.4

ed addresses ed

commands

Regular users of vi will be familiar with the ed commands w and q (write and quit). ed also recognises commands to delete lines of text, to replace characters with other characters and a number of other functions. Table 7.5 summarises some of the ed commands and their formats. In Table 7.5 range can match any of the address formats outlined in Table 7.4.

David Jones (20.01.00)

Page 180

85321, Systems Administration

Chapter 7: Text Manipulation

Address
linea

Purpose the append command, allows the user to add text after line number line the delete command, delete the lines specified by range and count and place them into the buffer buffer the join command, takes the lines
specified by range and count and makes them one line

range d buffer count

range j count

q line r file

quit the read command, read the contents

of the file file and place them after the line line

sh range s/RE/characters/options

start up a new shell the substitute command, find any

characters that match RE and replace them with characters but only in the range specified by range

u range w file

the undo command, the write command, write to the file file all the lines specified by range
Table 7.5

ed commands

For example
Some more examples of ed commands include
5,10s/hello/HELLO/

replace the first occurrence of hello with HELLO for all lines between 5 and 10
5,10s/hello/HELLO/g

replace all occurrences of hello with HELLO for all lines between 5 and 10
1,$s/^\(.\{20,20\}\)\(.*\)$/\2\1/

for all lines in the file, take the first 20 characters and put them at the end of the line

David Jones (20.01.00)

Page 181

85321, Systems Administration

Chapter 7: Text Manipulation

The last example The last example deserves a bit more explanation. Lets break it down into its components is the range for the command. In this case it is the whole file (from line 1 to the last line). The command is substitute so we are going to replace some text with some other text.
/^ 1,$s The 1,$

The / indicates the start of the RE. The ^ is a RE pattern and it is used to match the start of a line (see Table 7.2).
\(.\{20,20\}\)

This RE fragment .\{20,20\} will match any 20 characters. By surrounding it with \( \) those 20 characters will be stored in register 1. match any number of characters and surrounding it with \( \) means those characters will be placed into the next available register (register 2). The $ is the RE character that matches the end of the line. So this fragment takes all the characters after the first 20 until the end of the line and places them into register 2.
/\2\1/ \(.*\)$ The .* says

This specifies what text should replace the characters matched by the previous RE. In this case the \2 and the \1 refer to registers 1 and 2. Remember from above that the first 20 characters on the line have been placed into register 1 and the remainder of the line into register 2.

The sed command

is a non-interactive version of ed. sed is given a sequence of ed commands and then performs those commands on its standard input or on files passes as parameters. It is an extremely useful tool for a Systems Administrator. The ed and vi commands are interactive which means they require a human being to perform the tasks. On the other had sed is noninteractive and can be used in shell programs which means tasks can be automated.
sed

sed command format

By default the sed command acts like a filter. It takes input from standard input and places output onto standard output. sed can be run using a number of different formats.
sed command [file-list] sed [-e command] [-f command_file] [filelist]

is one of the valid ed commands. The -e command option can be used to specify multiple sed commands. For example,
command
David Jones (20.01.00) Page 182

85321, Systems Administration

Chapter 7: Text Manipulation

sed e '1,$s/david/DAVID/' e '1,$s/bash/BASH/' /etc/passwd

The -f command_file tells sed to take its commands from the file command_file. That file will contain ed commands one to a line. For example Some of the tasks you might use sed for include change the username DAVID in the /etc/passwd to david for any users that are currently using bash as their login shell change them over to the csh.

You could also use vi or ed to perform these same tasks. Note how the / in and /bin/csh have been quoted. This is because the / character is used by the substitute command to split the text to find and the text to replace it with. It is necessary to quote the / character so ed will treat it as a normal character.
/bin/bash sed 's/DAVID/david/' /etc/passwd sed 's/david/DAVID/' -e 's/\/bin\/bash/\/bin\/csh/' /etc/passwd sed -f commands /etc/passwd

The last example assumes that there is a file called commands that contains the following
s/david/DAVID/ s/\/bin\/bash/\/bin\/csh/

Understanding complex commands

When you combine regular expressions with ed commands you can get quite a long string of nearly incomprehensible characters. This can be quite difficult especially when you are just starting out with regular expressions. The secret to understanding these strings, like with many other difficult tasks, is breaking it down into smaller components. In particular, you need to learn to read the regular expression from the left to the right and understand each character as you go. For example, lets take the second substitute command from the last section
s/\/bin\/bash/\/bin\/csh/

We know it is an ed command so the first few characters are going to indicate what type of command. Going through the characters
s

The first character is an s followed by a / so that indicates a substitute command. Trouble is we dont know what the range is because it isnt specified. For most commands there will be a default value for the range. For the case of sed the default range is the current line.
/

In this position it indicates the start of the string the substitute command will search for.

David Jones (20.01.00)

Page 183

85321, Systems Administration

Chapter 7: Text Manipulation

We are now in the RE specifying the string to match. The \ is going to remove the special meaning from the next character.
/

Normally this would indicate the end of the string to match. However, the previous character has removed that special meaning. Instead we now know the first character we are matching a /
bin

Ive placed these together as they are normal characters. We are now trying to match /bin
\/

As before the \ removes the special meaning. So we are trying to match /bin/
bash

Now matching /bin/bash

/

No \ to remove the special meaning. So this indicates the end of the string to search for and the start of the replace string. Hopefully you have the idea by now and complete this process. This command will search for the string /bin/bash and replace it with /bin/csh Exercises
7.7.

Perform the following tasks with both vi and sed. You have just written a history of the UNIX operating system but you referred to UNIX as unix throughout. Replace all occurrences of unix with UNIX Youve just written a Pascal procedure using Write instead of Writeln. The procedure is part of a larger program. Replace Write with Writeln for all lines between the next occurrence of BEGIN and the following END When you forward a mail message using the elm mail program it automatically adds > to the beginning of every line. Delete all occurrences of > that start a line. What do the following ed commands do?
.+1,$d 1,$s/OSF/Open Software Foundation/g 1,/end/s/\([a-z]*\) \([0-9]*\)/\2 \1/

7.8.

7.9.

What are the following commands trying to do? Will they work? If not why not?
sed e 1,$s/^:/fred:/g /etc/passwd sed '1,$s/david/DAVID/' '1,$s/bash/BASH/' /etc/passwd

David Jones (20.01.00)

Page 184

85321, Systems Administration

Chapter 7: Text Manipulation

Conclusions
Regular expressions (REs) are a powerful mechanism for matching patterns of characters. REs are understood by a number of commands including vi, grep, sed, ed, awk and Perl. vi is just one of a family of editors starting with ed and including ex and sed. This entire family recognise ed commands that support the use of regular expressions to manipulate text.

Review Questions
7.1 Use vi and awk to perform the following tasks with the file 85321.txt (the student numbers have been changed to protect the innocent). This file is available from the 85321 Web site/CD-ROM under the resource materials section for week 3. Unless specified assume each task starts with the original file. remove the student number
y

switch the order for first name, last name remove any student with the name david

David Jones (20.01.00)

Page 185

85321, Systems Administration

Chapter 8: Shell Programming

Chapter
Shell Programming

Introduction
Shell Programming - WHY?
While it is very nice to have a shell at which you can issue commands, have you had the feeling that something is missing? Do you feel the urge to issue multiple commands by only typing one word? Do you feel the need for variables, logic conditions and loops? Do you strive for automation? If so, then welcome to shell programming. (If you answered no to any of the above then you are obviously in the wrong frame of mind to be reading this - please try again later :) Shell programming allows system administrators (and users) to create small (and occasionally not-so-small) programs for various purposes including automation of system administration tasks, text processing and installation of software. Perhaps the most important reason why a Systems Administrator needs to be able to read and understand shell scripts is the UNIX startup process. UNIX uses a large number of shell scripts to perform a lot of necessary system configuration when the computer first starts. If you cant read shell scripts you cant modify or fix the startup process.

Shell Programming - WHAT?

A shell program (sometimes referred to as a shell script) is a text file containing shell and UNIX commands. Remember - a UNIX command is a physical program (like cat, cut and grep) where as a shell command is an interpreted command - there isn't a physical file associated with the command; when the shell sees the command, the shell itself performs certain actions (for example, echo) When a shell program is executed the shell reads the contents of the file line by line. Each line is executed as if you were typing it at the shell prompt. There isn't anything that you can place in a shell program that you can't type at the shell prompt. Shell programs contain most things you would expect to find in a simple programming language. Programs can contain services including: variables logic constructs (IF THEN AND OR etc) looping constructs (WHILE FOR)

David Jones (20.01.00)

Page 186

85321, Systems Administration

Chapter 8: Shell Programming

functions comments (strangely the most least used service)

The way in which these services are implemented is dependant on the shell that is being used (remember - there is more than one shell). While the variations are often not major it does mean that a program written for the bourne shell (sh/bash) will not run in the c shell (csh). All the examples in this chapter are written for the bourne shell.

Shell Programming - HOW?

Shell programs are a little different from what youd usually class as a program. They are plain text and they dont need to be compiled. The shell "interprets" shell programs - the shell reads the shell program line by line and executes the commands it encounters. If it encounters an error (syntax or execution), it is just as if you typed the command at the shell prompt - an error is displayed. This is in contrast to C/C++, Pascal and Ada programs (to name but a few) which have source in plain text, but require compiling and linking to produce a final executable program. So, what are the real differences between the two types of programs? At the most basic level, interpreted programs are typically quick to write/modify and execute (generally in that order and in a seemingly endless loop :). Compiled programs typically require writing, compiling, linking and executing, thus are generally more time consuming to develop and test. However, when it comes to executing the finished programs, the execution speeds are often widely separated. A compiled/linked program is a binary file containing a collection direct systems calls. The interpreted program, on the other hand, must first be processed by the shell which then converts the commands to system calls or calls other binaries - this makes shell programs slow in comparison. In other words, shell programs are not generally efficient on CPU time. Is there a happy medium? Yes! It is called Perl. Perl is an interpreted language but is interpreted by an extremely fast, optimised interpreter. It is worth noting that a Perl program will be executed inside one process, whereas a shell program will be interpreted from a parent process but may launch many child processes in the form of UNIX commands (ie. each call to a UNIX command is executed in a new process). However, Perl is a far more difficult (but extremely powerful) tool to learn - and this chapter is called "Shell Programming"...

The Basics
A Basic Program
It is traditional at this stage to write the standard "Hello World" program. To do this in a shell program is so obscenely easy that were going to examine

David Jones (20.01.00)

Page 187

85321, Systems Administration

Chapter 8: Shell Programming

something a bit more complex - a hello world program that knows who you are... To create your shell program, you must first edit a file - name it something like "hello", "hello world" or something equally as imaginative - just dont call it "test" - we will explain why later. In the editor, type the following (or you could go to the 85321 website/CDROM and cut and paste the text from the appropriate web page)
#!/bin/bash # This is a program that says hello echo "Hello $LOGNAME, I hope you have a nice day!"

(You may change the text of line three to reflect your current mood if you wish) Now, at the prompt, type the name of your program - you should see something like:
bash: ./helloworld: Permission denied

Why? The reason is that your shell program isnt executable because it doesnt have its execution permissions set. After setting these (Hint: something involving the chmod command), you may execute the program by again typing its name at the prompt. An alternate way of executing shell programs is to issue a command at the shell prompt to the effect of:
<shell> <shell program>

eg
bash helloworld

This simply instructs the shell to take a list of commands from a given file (your shell script). This method does not require the shell script to have execute permissions. However, in general you will execute your shell scripts via the first method. And yet you may still find your script won't execute - why? On some UNIX systems (Red Hat Linux included) the current directory (.) is not included in the PATH environment variable. This mans that the shell can't find the script that you want to execute, even when it's sitting in the current directory! To get around this either: Modify the PATH variable to include the . directory:

PATH=$PATH:. Or, execute the program with an explicit path: ./helloworld

David Jones (20.01.00)

Page 188

85321, Systems Administration

Chapter 8: Shell Programming

An Explanation of the Program

Line one, #!/bin/bash is used to indicate which shell the shell program is to be run in. If this program was written for the C shell, then you might have #!/bin/csh instead. It is probably worth mentioning at this point that UNIX executes programs by first looking at the first two bytes of the file (this is similar to the way MSDOS looks at the first two bytes of executable programs; all .EXE programs start with MZ). From these two characters, the system knows if the file is an interpreted script (#!) or some other file type (more information can be obtained about this by typing man file). If the file is an interpreted script, then the system looks for a following path indicating an interpreter. For example:
#!/bin/bash #!/usr/bin/perl #!/bin/sh

Are all valid interpreters. Line two, # This is a program that says hello , is (you guessed it) a comment. The "#" in a shell script is interpreted as "anything to the right of this is a comment, go onto the next line". Note that it is similar to line one except that line one has the "!" mark after the comment. Comments are a very important part of any program - it is a really good idea to include some. The reasons why are standard to all languages - readability, maintenance and self congratulation. It is more so important for a system administrator as they very rarely remain at one site for their entire working career, therefore, they must work with other people's shell scripts (as other people must work with theirs). Always have a comment header; it should include things like:
# # # # # # # # # # # # # AUTHOR: DATE: PROGRAM: USAGE: PURPOSE: Who wrote it Date first written Name of the program How to run the script; include any parameters Describe in more than three words what the program does Files the shell script uses Optional but can include a list of "features" to be fixed Revisions/Changes

FILES: NOTES:

HISTORY:

This format isn't set in stone, but use common sense and write fairly self documenting programs.

David Jones (20.01.00)

Page 189

85321, Systems Administration

Chapter 8: Shell Programming

Version Control Systems Those of you studying software engineering may be familiar with the term, version control. Version control allows you to keep copies of files including a list of who made what changes and what those changes were. Version control systems can be very useful for keeping track of source code and is just about compulsory for any large programming project. Linux comes with CVS (Concurrent Versions System) a widely used version control system. While version control may not seem all that important it can save a lot of heartache. Many large sites will actually keep copies of system configuration files in a version control system. Line three, echo "Hello $LOGNAME, I hope you have a nice day!" is actually a command. The echo command prints text to the screen. Normal shell rules for interpreting special characters apply for the echo statement, so you should generally enclose most text in "". The only tricky bit about this line is the $LOGNAME . What is this? $LOGNAME is a shell variable; you can see it and others by typing "set" at the shell prompt. In the context of our program, the shell substitutes the $LOGNAME value with the username of the person running the program, so the output looks something like:
Hello jamiesob, I hope you have a nice day!

All variables are referenced for output by placing a "$" sign in front of them we will examine this in the next section. Exercises
8.1.

Modify the helloworld program so its output is something similar to:

Hello <username>, welcome to <machine name>

All You Ever Wanted to Know About Variables

You have previously encountered shell variables and the way in which they are set. To quickly revise, variables may be set at the shell prompt by typing:
[david@faile david]$ variable="a string"

Since you can type this at the prompt, the same syntax applies within shell programs. You can also set variables to the results of commands, for example:
[david@faile david]$ variable=ls -al

(Remember - the is the execute quote)

David Jones (20.01.00)

Page 190

85321, Systems Administration

Chapter 8: Shell Programming

To print the contents of a variable, simply type:

[david@faile david]$ echo $variable

Note that weve added the "$" to the variable name. Variables are always accessed for output with the "$" sign, but without it for input/set operations. Returning to the previous example, what would you expect to be the output? You would probably expect the output from ls -al to be something like:
drwxr-xr-x drwxr-xr-x -rw-r--r--rw-r--r--rw-r--r--rw-r--r-2 45 1 1 1 1 jamiesob jamiesob jamiesob jamiesob jamiesob jamiesob users users users users users users 1024 2048 851 12517 8 46604 Feb Feb Feb Feb Feb Feb 27 25 25 25 26 25 19:05 20:32 19:37 19:36 22:50 19:34 ./ ../ conX confile helloworld net-acct

and therefore, printing a variable that contains the output from that command would contain something similar, yet you may be surprised to find that it looks something like:
drwxr-xr-x 2 jamiesob users 1024 Feb 27 19:05 ./ drwxr-xr-x 45 jamiesob users 2048 Feb 25 20:32 ../ -rw-r--r-- 1 jamiesob users 851 Feb 25 19:37 conX -rw-r--r-- 1 jamiesob users 12517 Feb 25 19:36 confile -rw-r--r-- 1 jamiesob users 8 Feb 26 22:50 helloworld -rw-r-r-- 1 jamiesob users 46604 Feb 25 19:34 net-acct

Why?
When placing the output of a command into a shell variable, the shell removes all the end-of-line markers, leaving a string separated only by spaces. The use for this will become more obvious later, but for the moment, consider what the following script will do:
#!/bin/bash $filelist=ls cat $filelist

Exercise
8.2.

Type in the above program and run it. Explain what is happening. Would the above program work if "ls -al" was used rather than "ls" - Why/why not?

Predefined Variables
There are many predefined shell variables, most established during your login. Examples include $LOGNAME, $HOSTNAME and $TERM - these names are not always standard from system to system (for example, $LOGNAME can also be called $USER). There are however, several standard predefined shell variables you should be familiar with. These include:
$$ $? (The current process ID) (The exits status of last command)

How would these be useful?

David Jones (20.01.00)

Page 191

85321, Systems Administration

Chapter 8: Shell Programming

$$ $$ is extremely useful in creating unique temporary files. You will often find the following in shell programs:
some command > /tmp/temp.$$ . . some commands using /tmp/temp.$$> . . rm /tmp/temp.$$

/tmp/temp.$$ would always be a unique file - this allows several people to run the same shell script simultaneously. Since one of the only unique things about a process is its PID (Process-Identifier), this is an ideal component in a temporary file name. It should be noted at this point that temporary files are generally located in the /tmp directory. $? $? becomes important when you need to know if the last command that was executed was successful. All programs have a numeric exit status - on UNIX systems 0 indicates that the program was successful, any other number indicates a failure. We will examine how to use this value at a later point in time. Is there a way you can show if your programs succeeded or failed? Yes! This is done via the use of the exit command. If placed as the last command in your shell program, it will enable you to indicate, to the calling program, the exit status of your script. exit is used as follows:
exit 0 exit 1 # Exit the script, $? = 0 (success) # Exit the script, $? = 1 (fail)

Another category of standard shell variables are shell parameters.

Parameters - Special Shell Variables

If you thought shell programming was the best thing since COBOL, then you havent even begun to be awed - shell programs can actually take parameters. Table 8.1 lists each variable associated with parameters in shell programs:

David Jones (20.01.00)

Page 192

85321, Systems Administration

Chapter 8: Shell Programming

Variable $0 $1 thru $9 $# $* $@

Purpose the name of the shell program the first thru to ninth parameters the number of parameters all the parameters passed represented as a single word with individual parameters separated all the parameters passed with each parameter as a separate word
Shell Table 8.1 Parameter Variables

The following program demonstrates a very basic use of parameters:

#!/bin/bash # FILE: parm1 VAL=expr ${1:-0} + ${2:-0} + ${3:-0} echo "The answer is $VAL"

Pop Quiz: Why are we using ${1:-0} instead of $1? Hint: What would happen if any of the variables were not set? A sample testing of the program looks like:
[david@faile david]$ parm1 2 3 5 The answer is 10 [david@faile david]$ parm1 2 3 The answer is 5 [david@faile david]$ parm The answer is 0

Consider the program below:

#!/bin/bash # FILE: mywc

FCOUNT=ls $* 2> /dev/null | wc -w echo "Performing word count on $*" echo wc -w $* 2> /dev/null echo echo "Attempted to count words on $# files, found $FCOUNT"

If the program that was run in a directory containing:

conX confile helloworld my_file mywc* net-acct netnasties netnasties~ netwatch netwatch~ notes.txt notes.txt~ scanit* scanit~ shellprog shellprog~ study.htm study.txt study_~1.htm parm1* t1~ ttt tes/

Some sample testing would produce:

[david@faile david]$ mywc mywc Performing word count on mywc

David Jones (20.01.00)

Page 193

85321, Systems Administration

Chapter 8: Shell Programming

34 mywc Attempted to count words on 1 files, found [david@faile david]$ mywc mywc anotherfile Performing word count on mywc anotherfile 34 mywc 34 total Attempted to count words on 2 files, found 1 1

Exercise
8.3.

Explain line by line what this program is doing. What would happen if the user didnt enter any parameters? How could you fix this?

Only Nine Parameters?

Well thats what it looks like doesnt it? We have $1 to $9 - what happens if we try to access $10? Try the code below:
#!/bin/bash # FILE: testparms echo "$1 $2 $3 $4 $5 $6 $7 $8 $9 $10 $11 $12" echo $* echo $#

Run testparms as follows:

[david@faile david]$ testparms a b c d e f g h I j k l

The output will look something like:

a b c d e f g h i a0 a1 a2 a b c d e f g h I j k l 12

Why? The shell only has 9 command-line parameters defined at any one time $1 to $9. When the shell sees "$10" it interprets this as "$1" with a "0" after it. This is where $10 in the above results in a0. The a is the value of $1 with the 0 added. On the otherhand $* allows you to see all the parameters you typed! So how do you access $10, $11 etc. To our rescue comes the shift command. shift works by removing the first parameter from the parameter list and shuffling the parameters along. Thus $2 becomes $1, $3 becomes $2 etc. Finally, (what was originally) the tenth parameter becomes $9. However, beware! Once youve run shift, you have lost the original value of $1 forever - it is also removed from $* and $@. shift is executed by, well, placing the word "shift" in your shell script, for example:
#!/bin/bash echo $1 $2 $3

David Jones (20.01.00)

Page 194

85321, Systems Administration

Chapter 8: Shell Programming

shift echo $1 $2 $3

Exercise
8.4.

Modify the testparms program so the output looks something like: a b c d e f g h i a0 a1 a2 abcdefghIjkl 12 b c d e f g h i j b1 b2 b3 bcdefghijkl 11 c d e f g h i j k c0 c1 c2 cdefghIjkl 10

The difference between $* and $@

$* and $@ are very closely related. They both are expanded to become a list of all the command line parameters passed to a script. However, there are some subtle differences in how these two variables are treated. The subtleties are made even more difficult when they appear to act in a very similar way (in some situations). For example, lets see what happens with the following shell script
#for name in $* for name in $@ do echo param is $name done

The idea with this script is that you can test it with either $* or $@ by uncommenting the one you want to experiment and comment out the other line. The following examples show what happens when I run this script. The first time with $@, the second with $*
[david@faile david]$ tmp.sh hello "how are you" today 1 2 3 param is hello param is how param is are param is you param is today param is 1 param is 2 param is 3 [david@faile david]$ tmp.sh hello "how are you" today 1 2 3 param is hello param is how param is are param is you param is today param is 1 param is 2 param is 3

David Jones (20.01.00)

Page 195

85321, Systems Administration

Chapter 8: Shell Programming

As you can see no difference!! So whats all this fuss with $@ and $*? The difference comes when $@ and $* are used within double quotes. In this situation they work as follows
$@

Is expanded to all the command-line parameters joined as a single word with usually a space seperating them (the separating character can be changed).
$*

Expands to all the command-line parameters BUT each command-line parameter is treated as if it is surrounded by double quotes "". This is especially important when one of the parameters contains a space. Lets modify the our example script so that $@ and $* are surrounded by ""
#for name in "$*" for name in "$@" do echo param is $name done

Now look at what happens when we run it using the same parameters as before. Again the $@ version is executed first then the $* version.
[david@faile david]$ tmp.sh hello "how are you" today 1 2 3 param is hello param is how are you param is today param is 1 param is 2 param is 3 [david@faile david]$ tmp.sh hello "how are you" today 1 2 3 param is hello how are you today 1 2 3

With the second example, where $* is used, the difference is obvious. The first example, where $@ is used, shows the advantage of $@. The second parameter is maintained as a single parameter.

The basics of input/output (IO)

We have already encountered the "echo" command, yet this is only the "O" part of IO - how can we get user input into our programs? We use the "read" command. For example:
#!/bin/bash # FILE: testread read X echo "You said $X"

The purpose of this enormously exciting program should be obvious. Just in case you were bored with the echo command. Table 8.2 shows a few backslash characters that you can use to brighten your shell scripts:

David Jones (20.01.00)

Page 196

85321, Systems Administration

Chapter 8: Shell Programming

Character
\a \b \c \n \r \t \v \\ \nnn

Purpose alert (bell) backspace dont display the trailing newline new line carriage return horizontal tab vertical tab backslash the character with ASCII number nnn (octal)
Table 8.2 echo backslash options

(type "man echo" to see this exact table :) To enable echo to interpret these backslash characters within a string, you must issue the echo command with a "-e" switch. You may also add a "-n" switch to stop echo printing a new-line at the end of the string - this is a good thing if you want to output a prompting string. For example:
#!/bin/bash # FILE: getname echo -n "Please enter your name: " read NAME echo "Your name is $NAME"

(This program would be useful for those with a very short memory) At the moment, weve only examined reading from STDIN (standard input a.k.a. the keyboard) and STDOUT (standard output a.k.a. the screen) - if we want to be really clever we can change this. What do you think the following does?
read X < afile

or what about
echo $X > anotherfile

If you said that the first read the contents of afile into a variable $X and the second wrote the value of $X to anotherfile youd almost be correct. The read operation will only read the first line (up to the end-of-line marker) from afile - it doesnt read the entire file. You can also use the ">>" and "<<" redirection operators.

David Jones (20.01.00)

Page 197

85321, Systems Administration

Chapter 8: Shell Programming

Exercises
8.5.

What would you expect:

read X << END would

do? What do you think $X would hold if the input was:

Dear Sir I have no idea why your computer blew up. Kind regards, me. END

And now for the hard bits

Scenario
So far we have been dealing with very simple examples - mainly due to the fact weve been dealing with very simple commands. Shell scripting was not invented so you could write programs that ask you your name then display it. For this reason, we are going to be developing a real program that has a useful purpose. We will do this section by section as we examine more shell programming concepts. While you are reading each section, you should consider how the information could assist in writing part of the program. The actual problem is as follows: Youve been appointed as a system administrator to an academic department within a small (anonymous) regional university. The previous system administrator left in rather a hurry after it was found that department's main server had being playing host to plethora of pornography, warez (pirate software) and documentation regarding interesting alternative uses for various farm chemicals. There is some concern that the previous sys admin wasnt the only individual within the department who had been availing themselves to such wonderful and diverse resources on the Internet. You have been instructed to identify those persons who have been visiting "undesirable" Internet sites and advise them of the department's policy on accessing inappropriate material (apparently there isn't one, but you've been advised to improvise). Ideally, you will produce a report of people accessing restricted sites, exactly which sites and the number of times they visited them. To assist you, a network monitoring program produces a datafile containing a list of users and sites they have accessed, an example of which is listed below:
FILE: netwatch jamiesob mucus.slime.com tonsloye xboys.funnet.com.fr tonsloye sweet.dreams.com root sniffer.gov.au jamiesob marvin.ls.tc.hk jamiesob never.land.nz

David Jones (20.01.00)

Page 198

85321, Systems Administration

Chapter 8: Shell Programming

jamiesob guppy.pond.cqu.edu.au tonsloye xboys.funnet.com.fr tonsloye www.sony.com janesk horseland.org.uk root www.nasa.gov tonsloye warez.under.gr tonsloye mucus.slime.com root ftp.ns.gov.au tonsloye xboys.funnet.com.fr root linx.fare.com root crackz.city.bmr.au janesk smurf.city.gov.au jamiesob mucus.slime.com jamiesob mucus.slime.com

After careful consideration (and many hours of painstaking research) a steering committee on the departments policy on accessing the internet has produced a list of sites that they have deemed "prohibited" - these sites are contained in a data file, an example of which is listed below:
FILE: netnasties mucus.slime.com xboys.funnet.com.fr warez.under.gr crackz.city.bmr.au

It is your task to develop a shell script that will fulfil these requirements (at the same time ignoring the privacy, ethics and censorship issues at hand :) (Oh, it might also be an idea to get Yahoo! to remove the link to your main server under the /Computers/Software/Hackz/Warez/Sites listing... ;)

if ... then ... maybe?

Shell programming provides the ability to test the exit status from commands and act on them. One way this is facilitated is:
if command then do other commands fi

You may also provide an "alternate" action by using the "if" command in the following format:
if command then do other commands else do other commands fi

And if you require even more complexity, you can issue the if command as:
if command then do other commands elif anothercommand do other commands fi

David Jones (20.01.00)

Page 199

85321, Systems Administration

Chapter 8: Shell Programming

To test these structures, you may wish to use the true and false UNIX commands. true always sets $? to 0 and false sets $? to 1 after executing. Remember: if tests the exit code of a command - it isnt used to compare values; to do this, you must use the test command in combination with the if structure - test will be discussed in the next section. What if you wanted to test the output of two commands? In this case, you can use the shells && and || operators. These are effectively "smart" AND and OR operators. The && works as follows:
command1 && command2

command2 will only be executed if command1 succeeds. The || works as follows:

command1 || command2

command2 will only be executed if command1 fails. These are sometimes referred to as "short circuit" operators in other languages. Given our problem, one of the first things we should do in our program is to check if our datafiles exist. How would we do this?
#!/bin/bash # FILE: scanit if ls netwatch && ls netnasties then echo "Found netwatch and netnasties!" else echo "Can not find one of the data files - exiting" exit 1 fi

Exercise
8.6.

Enter the code above and run the program. Notice that the output from the ls commands (and the errors) appear on the screen - this isnt a very good thing. Modify the code so the only output to the screen is one of the echo messages.

Testing Testing...
Perhaps the most useful command available to shell programs is the test command. It is also the command that causes the most problems for first time shell programmers - the first program they ever write is usually (imaginatively) called test - they attempt to run it - and nothing happens - why? (Hint: type which test, then type echo $PATH - why does the system command test run before the programmers shell script?) The test command allows you to: test the length of a string

David Jones (20.01.00)

Page 200

85321, Systems Administration

Chapter 8: Shell Programming

compare two strings compare two numbers check on a files type check on a files permissions combine conditions together

test actually comes in two flavours:

test an_expression

and
[ an_expression ]

They are both the same thing - its just that [ is soft-linked to /usr/bin/test ; test actually checks to see what name it is being called by; if it is [ then it expects a ] at the end of the expression. What do we mean by "expression"? The expression is the string you want evaluated. A simple example would be:
if [ "$1" = "hello" ] then echo "hello to you too!" else echo "hello anyway" fi

This simply tests if the first parameter was hello. Note that the first line could have been written as:
if test "$1" = "hello"

Tip: Note that we surrounded the variable $1 in quotes. This is to take care of the case when $1 doesnt exist - in other words, there were no parameters passed. If we had simply put $1 and there wasnt any $1, then an error would have been displayed:
test: =: unary operator expected

This is because youd be effectively executing:

test NOTHING = "hello"

= expects a string to its left and right - thus the error. However, when placed in double quotes, you be executing:
test "" = "hello"

which is fine; youre testing an empty string against another string. You can also use test to tell if a variable has a value in it by:
test $var

This will return true if the variable has something in it, false if the variable doesnt exist OR it contains null (""). We could use this in our program. If the user enters at least one username to check on, them we scan for that username, else we write an error to the screen and exit:

David Jones (20.01.00)

Page 201

85321, Systems Administration

Chapter 8: Shell Programming

if [ $1 ] then the_user_list=echo $* else echo "No users entered - exiting! exit 2 fi

Expressions, expressions!
So far weve only examined expressions containing string based comparisons. The following tables list all the different types of comparisons you can perform with the test command. Expression
-z string -n string string1 = string2 string != string2 String

True if length of string is 0 length of string is not 0 if the two strings are identical if the two strings are NOT identical if string is not NULL
Table 8.3 String based tests

Expression
int1 -eq int2 int1 -ne int2 int1 -gt int2 Int1 -ge int2 Int1 -lt int2 Int1 -le int2

True if first int is equal to second first int is not equal to second first int is greater than second first int is greater than or equal to second first int is less than second first int is less than or equal to second
Table 8.4 Numeric tests

David Jones (20.01.00)

Page 202

85321, Systems Administration

Chapter 8: Shell Programming

Expression
-r file -w file -x file -f file -d file -h file -c file -b file -p file -u file -g file -k file -s file File file file file file file file file file file file file file

True if exists and is readable exists and is writable exists and is executable exists and is a regular file exists and is directory exists and is a symbolic link exists and is a character special file exists and is a block special file exists and is a named pipe exists and it is setuid exists and it is setgid exists and the sticky bit is set exists and its size is greater than 0
Table 8.5 File tests

Expression
! -a -o ( expr )

Purpose reverse the result of an expression AND operator OR operator group an expression, parentheses have special meaning to the shell so to use them in the test command you must quote them
Logic Table 8.6 operators with test

Remember: test uses different operators to compare strings and numbers using -ne on a string comparison and != on a numeric comparison is incorrect and will give undesirable results.

Exercise
8.7.

Modify the code for scanit so it uses the test command to see if the datafiles exists.

All about case

Ok, so we know how to conditionally perform operations based on the return status of a command. However, like a combination between the if statement and the test $string = $string2, there exists the case statement.
David Jones (20.01.00) Page 203

85321, Systems Administration

Chapter 8: Shell Programming

case value in pattern 1) command anothercommand ;; pattern 2) command anothercommand ;; esac

case works by comparing value against the listed patterns. If a match is made, then the commands associated with that pattern are executed (up to the ";;" mark) and $? is set to 0. If a match isnt made by the end of the case statement (esac) then $? is set to 1. The really useful thing is that wildcards can be used, as can the "|" symbol which acts as an OR operator. The following example gets a Yes/No response from a user, but will accept anything starting with "Y" or "y" as YES, "N" or "n" as no and anything else as "MAYBE"
echo read case Y* N* *) esac echo -n "Your Answer: " ANSWER $ANSWER in | y*) ANSWER="YES" ;; | n*) ANSWER="NO" ;; ANSWER="MAYBE" ;; $ANSWER

Exercise
8.8.

Write a shell script that inputs a date and converts it into a long date form. For example: $~ > mydate 12/3/97 12th of March 1997 $~ > mydate Enter the date: 1/11/74 1st of November 1974

Loops and Repeated Action Commands

Looping - "the exciting process of doing something more than once" - and shell programming allows it. There are three constructs that implement looping:
while - do - done for - do - done until - do - done

while
The format of the while construct is:
while command do commands done

(while command is true, commands are executed)

David Jones (20.01.00)

Page 204

85321, Systems Administration

Chapter 8: Shell Programming

Example
while [ $1 ] do echo $1 shift done

What does this segment of code do? Try running a script containing this code with a b c d e on the command line. while also allows the redirection of input. Consider the following:
#!/bin/bash # FILE: linelist # count=0 while read BUFFER do count=expr $count + 1 # Increment the count echo "$count $BUFFER" # Echo it out done < $1 # Take input from the file

This program reads a file line by line and echos it to the screen with a line number. Given our scanit program, the following could be used read the netwatch datafile and compare the username with the entries in the datafile:
while read buffer do user=echo $buffer | cut -d" " -f1 site=echo $buffer | cut -d" " -f2 if [ "$user" = "$1" ] then echo "$user visited $site" fi done < netwatch

Exercise
8.9.

Modify the above code so that the site is compared with all sites in the prohibited sites file (netnasties). Do this by using another while loop. If the user has visited a prohibited site, then echo a message to the screen.

for
The format of the for construct is:
for variable in list_of_variables do commands done

(for each value in list_of_variables, "commands" are executed)

David Jones (20.01.00)

Page 205

85321, Systems Administration

Chapter 8: Shell Programming

Example We saw earlier in this chapter examples of the for command showing the difference between $* and $@. Another example
for count in 10 9 8 7 6 5 4 3 2 1 do echo -n "$count.." done echo

Modifying scanit Given our scanit program, we might wish to report on a number of users. The following modifications will allow us to accept and process multiple users from the command line:
for checkuser in $* do while read buffer do while read checksite do user=echo $buffer | cut -d" " -f1 site=echo $buffer | cut -d" " -f2 if [ "$user" = "$checkuser" -a "$site" = "$checksite" ] then echo "$user visited the prohibited site $site" fi done < netnasties done < netwatch done

Problems with running scanit

A student in the 1999 offering of 85321 reported the following problem with the scanit program on page 160 of chapter 8 of the 85321 textbook. When running her program she types
Bash scanit jamiesob

and quite contrary to expectations she gets 80 lines of output that includes
root visited the prohibited site crackz.city.bmr.au root visited the prohibited site crackz.city.bmr.au janesk visited the prohibited site smurf.city.gov.au janesk visited the prohibited site smurf.city.gov.au janesk visited the prohibited site smurf.city.gov.au janesk visited the prohibited site smurf.city.gov.au jamiesob visited the prohibited site mucus.slime.com jamiesob visited the prohibited site mucus.slime.com

If everything is working the output you should get is three lines of code reporting that the user jamiesob has visited the site mucus.slime.com. So what is the problem?
David Jones (20.01.00) Page 206

85321, Systems Administration

Chapter 8: Shell Programming

Well lets have a look at her shell program

for checkuser in $* do while read buffer do while read checksite do user=echo $buffer | cut -d" " -f1 site=echo $buffer | cut -d" " -f2 if [ "$user"="$checkuser" -a "$site"="$checksite" ] then echo "$user visited the prohibited site $site" fi done < netnasties done < netwatch done

Can you see the problem? How do we identify the problem? Well lets start by thinking about what the problem is. The problem is that it is showing too many lines. The script is not excluding lines which should not be displayed. Where are the lines displayed? The only place is within the if command. This seems to imply that the problem is that the if command isnt working. It is matching too many times, in fact it is matching all of the lines. The problem is that if command is wrong or not working as expected. How is it wrong? Common mistakes with the if command include not using the test command Some people try comparing "things" without using the test command
if "$user"="$checkuser" -a "$site"="$checksite"

The student is using the test command in our example. In fact, she is using the [ form of the test command. So this isnt the problem. using the wrong comparison operator Some people try things like
if [ "$user" == "$checkuser" ] or if [ "$user" -eq "$checkuser" ] Trouble with this is that == is comparison operator from the C/C++

programming languages and not a comparison operator supported by the test command. -eq is a comparison operator supported by test but it is used to compare numbers not strings. This isnt the problem here. The problem here is some missing spaces around the = signs. Remember that [ is actually a shell command (its the same command test). Like other commands it takes parameters. Lets have a look at the parameters that the test command takes in this example program. The test command is [ "$user"="$checkuser" -a "$site"="$checksite" ]

David Jones (20.01.00)

Page 207

85321, Systems Administration

Chapter 8: Shell Programming

Parameters must be surrounded by spaces. So this command has four parameters (the first [ is the command name)
1. "$user"="$checkuser" 2. -a 3. "$site"="$checksite" 4. ]

By now you might start to see the problem. For the test command to actual compare two "things" it needs to see the = as a separate parameter. The problem is that because there are no spaces around the = characters in this test command the = is never seen. Its just part of a string. The solution to this problem is to put some space characters around the two =. So we get
[ "$user" = "$checkuser" -a "$site" = "$checksite" ]

So what is happening
So what is actually happening? Why is the test always returning true. We know this because the script displays a line for all the users and all the sites. To find the solution to this problem we need to take a look at the manual page for the test command. On current Linux computers you can type man test and you will see a manual page for this command. However, it isnt the one you should look at. Type the following command which test. It should tell you where the executable program for test is located. Trouble is that on current Linux computers it wont. Thats because there isnt one. Instead the test command is actually provided by the shell, in this case bash. To find out about the test command you need to look at the man page for bash. The other approach would be to look at Table 8.3 from chapter 8 of the 85321 textbook. In particular the last entry which says that if the expression in a test command is a string then the test command will return true if the string is nonzero (i.e. it has some characters). Here are some examples to show what this actually means. In these examples Im using the test command by itself and then using the echo command to have a look at the value of the $? shell variable. The $? shell variable holds the return status of the previous command. For the test command if the return status is 0 then the expression was true. If it is 1 then the expression as false.
[david@faile [david@faile 0 [david@faile [david@faile 1 [david@faile [david@faile 0
David Jones (20.01.00)

8]$ [ fred ] 8]$ echo $? 8]$ [ ] 8]$ echo $? 8]$ [ "jamiesob"="mucus.slime.com" ] 8]$ echo $?

Page 208

85321, Systems Administration

Chapter 8: Shell Programming

[david@faile 8]$ [ "jamiesob" = "mucus.slime.com" ] [david@faile 8]$ echo $? 1

In the first example the expression is fred a string with a non-zero length. So the return status is 0 indicating true. In the second example there is no expression, so it is a string with zero length. So the return status is 1 indicating false. The last two examples are similar to the problem and solution in the students program. The third example is similar to the students problem. The parameter is a single non-zero length string ("jamiesob"="mucus.slime.com") so the return status is 0 indicating truth. When we add the spaces around the = we finally get what we wanted. The test command actually compares the two strings and sets the return status accordingly and because the strings are different the return status is 1 indicating false. So what about the -a operator used in the students program. Well the -a simply takes the results of two expressions (one on either side) and ands them together. In the students script there the two expressions are non-zero length strings. Which are always true. So that becomes 0 -a 0 (TRUE and TRUE) which is always true. Here are some more examples
[david@faile [david@faile 0 [david@faile [david@faile 1 [david@faile "david" ] [david@faile 1 [david@faile [david@faile 0 8]$ [ "jamiesob"="mucus.slime.com" -a "david"="fred" ] 8]$ echo $? 8]$ [ "jamiesob"="mucus.slime.com" -a "" ] 8]$ echo $? 8]$ [ "jamiesob" = "mucus.slime.com" -a "david" = 8]$ echo $? 8]$ [ "jamiesob" = "jamiesob" -a "david" = "david" ] 8]$ echo $?

The first example here is what is happening in the students program. Two nonzero length strings, which are always true, "anded" together will always return true regardless of the strings. The second example shows what happens when one side of the -a is a zero length string. A zero length string is always false, false and true is always false, so this example has a return status of 1 indicating false. The last two examples show "working" versions of the test command with spaces in all the right places. Where the two strings being compared are different the comparison is false and the test command is returning false. Where the two strings being compared are the same the comparison operator is returning true and the test command is returning true.

David Jones (20.01.00)

Page 209

85321, Systems Administration

Chapter 8: Shell Programming

Exercises
8.10.

What will be the return status of the following test commands? Why?
["hello"] [ $HOME ] [ "hello" ]

8.11.

The above code is very inefficient IO wise - for every entry in the netwatch file, the entire netnasties file is read in. Modify the code so that the while loop reading the netnasties file is replaced by a for loop. (Hint: what does: BADSITES=cat netnasties do?) EXTENSION: What other IO inefficiencies does the code have? Fix them.

Speed and shell scripts

Exercise 8.11 is actually a very important problem in that it highlights a common mistake made by many novice shell programmers. This mistake is especially prevalent amongst people who have experience in an existing programming language like C/C++ or Pascal. This supplementary material is intended to address that problem and hopefully make it a little easier for you to answer question 11. Online lecture 8, particularly on slide 21 also addresses this problem. You might want to have a look at and listen to this slide before going much further.

Whats the mistake

A common mistake for beginning shell programmers make is to write shell programs as if they were C/C++ programs. In particular they tend not to make use of the collection of very good commands which are available. Lets take a look at a simple example of what I mean. The problem is to count the number of lines in a file (the file is called the_file). The following section discusses three solutions to this problem: a solution in C, a shell solution written like the C program, and a "proper" shell/UNIX solution

Solution in C
#include <stdio.h> void main( void ) { int line_count = 0; FILE *infile; char line[500]; infile = fopen( "the_file", "r" ); while ( ! feof( infile ) ) { fgets( line, 500, infile ); line_count++; }

David Jones (20.01.00)

Page 210

85321, Systems Administration

Chapter 8: Shell Programming

printf( "Number of lines is %d\n", line_count-1 ); }

Pretty simple to understand? Open the file, read the file line by line, increment a variable for each line and then display the variable when we reach the end of the file.

Shell solution written by C programmer

It is common for new comers to the shell to write shell scripts like C (or whatever procedural language they are familiar with) programs. Heres a shell version of the previous C solution. It uses the same algorithm.
count=0 while read line do count=expr $count + 1 done < the_file echo Number of lines is $count

This shell script reads the file line by line, increment a variable for each line and when we reach the end of the file display the value.

Shell solution by shell programmer

Anyone with a modicum of UNIX experience will know that you dont need to write a shell program to solve this problem. You just use the wc command. wc -l the_file This may appear to be a fairly trivial example. However, it does emphasise a very important point. You dont want to use the shell commands like a normal procedural programming language. You want to make use of the available UNIX commands where ever possible.

Comparing the solutions

Lets compare the solutions. The C program is obviously the longest solution when it comes to size of the program. The shell script is much shorter. The shell takes care of a lot of tasks you have to do with C and the use of wc is by far the shortest. The UNIX solutions are also much faster to write as there is no need for a compile/test cycle. This is one of the advantages of scripting languages like the shell, Perl and TCL. What about speed of execution? As weve seen in earlier chapters you can test the speed of executable programs (in a very coarse way) with the time command. The following shows the time taken for each solution. In the tests each of the three solutions worked on the same file which contained 1911 lines.
[david@faile david]$ time ./cprogram Number of lines is 1911 0.00user 0.01system 0:00.01elapsed 83%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (79major+11minor)pagefaults 0swaps

David Jones (20.01.00)

Page 211

85321, Systems Administration

Chapter 8: Shell Programming

[david@faile david]$ time sh shsolution Number of lines is 1911 12.24user 14.17system 0:28.12elapsed 93%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (164520major+109070minor)pagefaults 0swaps [david@faile david]$ time wc -l /var/log/messages 1911 /var/log/messages 0.00user 0.01system 0:00.04elapsed 23%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (85major+14minor)pagefaults 0swaps

The lesson to draw from these figures is that solutions using the C program and the wc command have the same efficiency but using the wc command is much quicker. The shell programming solution which was written like a C program is horrendously inefficient. It is tens of thousands of times slower than the other two solutions and uses an enormous amount of resources.

The problem
Obviously using while loops to read a file line by line in a shell program is inefficient and should be avoided. However, if you think like a C programmer you dont know any different. When writing shell programs you need to modify how you program to make use of the strengths and avoid the weaknesses of shell scripting. Where possible you should use existing UNIX commands.

A solution for scanit?

Just because the current implementation of scanit uses two while loops it doesnt mean that your solution has to. Think about the problem you have to solve. In the case of improving the efficiency of scanit you have to do the following for every user entered as a command line parameter see if the user has visited one of the sites listed in the netnasties file

To word it another way, you are searching for lines in a file which match a certain criteria. What UNIX command does that?

Number of processes
Another factor to keep in mind is the number of processes your shell script creates. Every UNIX command in a shell script will create a new process. Creating a new process is quite a time and resource consuming job performed by the operating system. One thing you want to do is to reduce the number of new processes created. Lets take a look at the shell program solution to our problem
count=0 while read line do count=expr $count + 1 done < the_file

David Jones (20.01.00)

Page 212

85321, Systems Administration

Chapter 8: Shell Programming

echo Number of lines is $count

For a file with 1911 lines this shell program is going to create about 1913 processes. 1 process for the echo command at the end, one process to for a new shell to run the script and 1911 processes for the expr command. Every time the script reads a line it will create a new process to run the expr command. So the longer the file the less efficient this script is going to get. One way to address this problem somewhat is to use the support that the bash shell provides for arithmetic. By using the shells arithmetic functions we can avoid creating a new process because the shell process will do it. Our new shell script looks like this
count=0 while read line do count=$[ $count + 1 ] done < /var/log/messages echo Number of lines is $count

See the change in the line incrementing the count variable. Its now using the shell arithmetic support. Look what happens to the speed of execution.
[david@faile 8]$ time bash test6 Number of lines is 1915 1.28user 0.52system 0:01.83elapsed 98%CPU (0avgtext+0avgdata 0maxresident)k 0inputs+0outputs (179major+30minor)pagefaults 0swaps

We have a slightly bigger file but even so the speed is much, much better. However, the speed is still no where as good as simply using the wc command.

until
The format of the until construct is:
until command do commands done

("commands" are executed until "command" is true) Example

until [ "$1" = "" ] do echo $1 shift done

David Jones (20.01.00)

Page 213

85321, Systems Administration

Chapter 8: Shell Programming

break and continue

Occasionally you will want to jump out of a loop; to do this you need to use the break command. break is executed in the form:
break

or
break n

The first form simply stops the loop, for example:

while true do read BUFFER if [ "$BUFFER" = "" ] then break fi echo $BUFFER done

This code takes a line from the user and prints it until the user enters a blank line. The second form of break, break n (where n is a number) effectively works by executing break "n" times. This can break you out of embedded loops, for example:
for file in $* do while read BUFFER do if [ "$BUFFER" = "ABORT" ] then break 2 fi echo $BUFFER done < $file done

This code prints the contents of multiple files, but if it encounters a line containing the word "ABORT" in any one of the files, it stops processing. Like break, continue is used to alter the looping process. However, unlike break, continue keeps the looping process going; it just fails to finish the remainder of the current loop by returning to the top of the loop. For example:
while read BUFFER do charcount=echo $BUFFER | wc -c | cut -f1 if [ $charcount -gt 80 ] then continue fi echo $BUFFER done < $1

This code segment reads and echos the contents of a file - however, it does not print lines that are over 80 characters long.

David Jones (20.01.00)

Page 214

85321, Systems Administration

Chapter 8: Shell Programming

Redirection
Not just the while - do - done loops can have IO redirection; it is possible to perform piping, output to files and input from files on if, for and until as well. For example:
if true then read x read y read x fi < afile

This code will read the first three lines from afile. Pipes can also be used:
read BUFFER while [ "$BUFFER" != "" ] do echo $BUFFER read BUFFER done | todos > tmp.$$

This code uses a non-standard command called todos. todos converts UNIX text files to DOS textfiles by making the EOL (End-Of-Line) character equivalent to CR (Carriage-Return) LF (Line-Feed). This code takes STDIN (until the user enters a blank line) and pipes it into todos, which in turn converts it to a DOS style text file ( tmp.$$ ) . In all, a totally useless program, but it does demonstrate the possibilities of piping.

Now for the really hard bits

Functional Functions
A symptom of most usable programming languages is the existence of functions. Theoretically, functions provide the ability to break your code into reusable, logical compartments that are the by product of top-down design. In practice, they vastly improve the readability of shell programs, making it easier to modify and debug them. An alternative to functions is the grouping of code into separate shell scripts and calling these from your program. This isnt as efficient as functions, as functions are executed in the same process that they were called from; however other shell programs are launched in a separate process space - this is inefficient on memory and CPU resources. You may have noticed that our scanit program has grown to around 30 lines of code. While this is quite manageable, we will make some major changes later that really require the "modular" approach of functions. Shell functions are declared as:
function_name() { somecommands }

David Jones (20.01.00)

Page 215

85321, Systems Administration

Chapter 8: Shell Programming

Functions are called by:

function_name parameter_list

YES! Shell functions support parameters. $1 to $9 represent the first nine parameters passed to the function and $* represents the entire parameter list. The value of $0 isnt changed. For example:
#!/bin/bash # FILE: catfile() { for file in $* do cat $file done } FILELIST=ls $1 cd $1 catfile $FILELIST catfiles

This is a highly useless example (cat * would do the same thing) but you can see how the "main" program calls the function.

local
Shell functions also support the concept of declaring "local" variables. The local command is used to do this. For example:
#!/bin/bash testvars() { local localX="testvars localX" X="testvars X" local GlobalX="testvars GlobalX" echo "testvars: localX= $localX X= $X GlobalX= $GlobalX" } X="Main X" GlobalX="Main GLobalX" echo "Main 1: localX= $localX X= $X GlobalX= $GlobalX" testvars echo "Main 2: localX= $localX X= $X GlobalX= $GlobalX"

The output looks like:

Main 1: localX= X= Main X GlobalX= Main GLobalX testvars: localX= testvars localX X= testvars X GlobalX= testvars GlobalX Main 2: localX= X= testvars X GlobalX= Main GLobalX

David Jones (20.01.00)

Page 216

85321, Systems Administration

Chapter 8: Shell Programming

The return trip

After calling a shell function, the value of $? is set to the exit status of the last command executed in the shell script. If you want to explicitly set this, you can use the return command:
return n

(Where n is a number) This allows for code like:

if function1 then do_this else do_that fi

For example, we can introduce our first function into our scanit program by placing our datafile tests into a function:
#!/bin/bash # FILE: # scanit

check_data_files() { if [ -r netwatch -a -r netnasties ] then return 0 else return 1 fi } # Main Program if check_data_files then echo "Datafiles found" else echo "One of the datafiles missing - exiting" exit 1 fi # our other work...

Difficult and not compulsory

The following section (up to the section titled "Bugs and Debugging") is not compulsory for students studying 85321.

Recursion: (see "Recursion")

Shell programming even supports recursion. Typically, recursion is used to process tree-like data structures - the following example illustrates this:

David Jones (20.01.00)

Page 217

85321, Systems Administration

Chapter 8: Shell Programming

#!/bin/bash # FILE:

wctree

wcfiles() { local BASEDIR=$1 # Set the local base directory local LOCALDIR=pwd # Where are we? cd $BASEDIR # Go to this directory (down) local filelist=ls # Get the files in this directory for file in $filelist do if [ -d $file ] # If we are a directory, recurs then # we are a directory wcfiles "$BASEDIR/$file" else fc=wc -w < $file # do word count and echo info echo "$BASEDIR/$file $fc words" fi done cd $LOCALDIR # Go back up to the calling directory } if [ $1 ] then wcfile $1 else wcfile "." fi # Default to . if no parms

Exercise
8.12.

What does the wctree program do? Why are certain variables declared as local? What would happen if they were not? Modify the program so it will only "recurs" 3 times. EXTENSION: There is actually a UNIX command that will do the same thing as this shell script - what is it? What would be the command line? (Hint: man find)

waiting and traping

So far we have only examined linear, single process shell script examples. What if you want to have more than one action occurring at once? As you are aware, it is possible to launch programs to run in the background by placing an ampersand behind the command, for example:
runcommand &

You can also do this in your shell programs. It is occasionally useful to send a time consuming task to the background and proceed with your processing. An example of this would be a sort on a large file:
sort $largefile > $newfile & do_a_function do_another_funtion $newfile

David Jones (20.01.00)

Page 218

85321, Systems Administration

Chapter 8: Shell Programming

The problem is, what if the sort hadnt finished by the time you wanted to use $newfile? The shell handles this by providing wait :
sort $largefile > $newfile & do_a_function wait do_another_funtion $newfile

When wait is encountered, processing stops and "waits" until the child process returns, then proceeds on with the program. But what if you had launched several processes in the background? The shell provides the shell variable $! (the PID of the child process launched) which can be given as a parameter to wait - effectively saying "wait for this PID". For example:
sort $largefile1 > $SortPID1=$! sort $largefile2 > $SortPID2=$! sort $largefile3 > $SortPID3=$! do_a_function wait $SortPID1 do_another_funtion wait $SortPID2 do_another_funtion wait $SortPID3 do_another_funtion $newfile1 & $newfile2 & $newfile3 &

$newfile1 $newfile2 $newfile3

Another useful command is trap. trap works by associating a set of commands with a signal from the operating system. You will probably be familiar with:
kill -9 PID

which is used to kill a process. This command is in fact sending the signal "9" to the process given by PID. Available signals are shown in Table 8.7. While you cant actually trap signal 9, you can trap the others. This is useful in shell programs when you want to make sure your program exits gracefully in the event of a shutdown (or some such event) (often you will want to remove temporary files the program has created). The syntax of using trap is:
trap commands signals

For example:
trap "rm /tmp/temp.$$" 1 2

will trap signals 1 and 2 - whenever these signals occur, processing will be suspended and the rm command will be executed. You can also list every traped signal by issuing the command:
trap

To "un-trap" a signal, you must issue the command:

trap "" signals

David Jones (20.01.00)

Page 219

85321, Systems Administration

Chapter 8: Shell Programming

Signal 0 1 2 3 4 5 6 7 8 10 12 13 14 15

Meaning Exit from the shell Hangup Interrupt Quit Illegal Instruction Trace trap IOT instruction EMT instruction Floating point exception Bus error Bad argument Pipe write error Alarm Software termination signal
Table 8.7 UNIX signals

(Taken from "UNIX Shell Programming" Kochan et al) The following is a somewhat clumsy form of IPC (Inter-Process Communication) that relies on trap and wait:
#!/bin/bash # FILE: saymsg # USAGE: saymsg <create number of children> [total number of # children] readmsg() { read line < $$ # read a line from the file given by the PID echo "$ID - got $line!" # of my *this* process ($$) if [ $CHILD ] then writemsg $line # if I have children, send them message fi } writemsg() { echo $* > $CHILD kill -1 $CHILD } stop() { kill -15 $CHILD if [ $CHILD ]

# Write line to the file given by PID # of my child. Then signal the child.

# tell my child to stop

David Jones (20.01.00)

Page 220

85321, Systems Administration

Chapter 8: Shell Programming

then wait $CHILD rm $CHILD fi exit 0 }

# wait until they are dead # remove the message file

# Main Program if [ $# -eq 1 ] then NUMCHILD=expr $1 - 1 saymsg $NUMCHILD $1 & # Launch another child CHILD=$! ID=0 touch $CHILD # Create empty message file echo "I am the parent and have child $CHILD" else if [ $1 -ne 0 ] # Must I create children? then NUMCHILD=expr $1 - 1 # Yep, deduct one from the number saymsg $NUMCHILD $2 & # to be created, then launch them CHILD=$! ID=expr $2 - $1 touch $CHILD # Create empty message file echo "I am $ID and have child $CHILD" else ID=expr $2 - $1 # I dont need to create children echo "I am $ID and am the last child" fi fi trap "readmsg" 1 trap "stop" 15 # Trap the read signal # Trap the drop-dead signal

if [ $# -eq 1 ] # If I have one parameter, then # then I am the parent - I just read read BUFFER # STDIN and pass the message on while [ "$BUFFER" ] do writemsg $BUFFER read BUFFER done echo "Parent - Stopping" stop else # Else I am the child who does nothing while true # I am totally driven by signals. do true done fi

So what is happening here? It may help if you look at the output:

psyche:~/sanotes[david@faile david]$ saymsg 3 I am the parent and have child 8090 I am 1 and have child 8094 I am 2 and have child 8109 I am 3 and am the last child this is the first thing I type 1 - got this is the first thing I type!

David Jones (20.01.00)

Page 221

85321, Systems Administration

Chapter 8: Shell Programming

2 - got this is the first thing I type! 3 - got this is the first thing I type! Parent - Stopping psyche:~/sanotes[david@faile david]$

Initially, the parent program starts, accepting a number of children to create. The parent then launches another program, passing it the remaining number of children to create and the total number of children. This happens on every launch of the program until there are no more children to launch. From this point onwards the program works rather like Chinese whispers - the parent accepts a string from the user which it then passes to its child by sending a signal to the child - the signal is caught by the child and readmsg is executed. The child writes the message to the screen, then passes the message to its child (if it has one) by signalling it and so on and so on. The messages are passed by being written to files - the parent writes the message into a file named by the PID of the child process. When the user enters a blank line, the parent process sends a signal to its child - the signal is caught by the child and stop is executed. The child then sends a message to its child to stop, and so on and so on down the line. The parent process cant exit until all the children have exited. This is a very contrived example - but it does show how processes (even at a shell programming level) can communicate. It also demonstrates how you can give a function name to trap (instead of a command set). Exercise
8.13.

saymsg is riddled with problems - there isnt any checking on the parent process command line parameters (what if there wasnt any?) and it isnt very well commented or written - make modifications to fix these problems. What other problems can you see? EXTENSION: Fundamentally saymsg isnt implementing very safe inter-process communication - how could this be fixed? Remember, one of the main problems concerning IPC is the race condition - could this happen?

Bugs and Debugging

If by now you have typed every example program in, completed every exercise and have not encountered one single error then you are a truly amazing person. However, if you are like me, you would have made at least 70 billion mistakes/ typos or TSEs (totally stupid errors) - and now I tell you the easy way to find them!

David Jones (20.01.00)

Page 222

85321, Systems Administration

Chapter 8: Shell Programming

Method 1 - set
Issuing the truly inspired command of:
set -x

within your program will do wonderful things. As your program executes, each code line will be printed to the screen - that way you can find your mistakes, err, well, a little bit quicker. Turning tracing off is a good idea once your program works - this is done by:
set +x

Method 2 - echo
Placing a few echo statements in your code during your debugging is one of the easiest ways to find errors - for the most part this will be the quickest way of detecting if variables are being set correctly.

Very Common Mistakes

$VAR=ls

This should be VAR=ls. When setting the value of a shell variable you dont use the $ sign.
read $BUFFER

The same thing here. When setting the value of a variable you dont use the $ sign.
VAR=ls -al"

The second is missing

if [ $VAR ] then echo $VAR fi

Havent specified what is being tested here. Need to refer to the contents of Tables 8.2 through 8.5
if [ $VAR -eq $VAR2 ] then echo $VAR fi

If $VAR and $VAR2 are strings then you cant use eq to compare their values. You need to use =.
if [ $VAR = $VAR2 ] then echo $VAR fi

The then must be on a separate line.

David Jones (20.01.00)

Page 223

85321, Systems Administration

Chapter 8: Shell Programming

And now for the really really hard bits

Writing good shell programs
We have covered most of the theory involved with shell programming, but there is more to shell programming than syntax. In this section, we will complete the scanit program, examining efficiency and structure considerations. scanit currently consists of one chunk of code with one small function. In its current form, it doesnt meet the requirements specified: "...you will produce a report of people accessing restricted sites, exactly which sites and the number of times they visited them." Our code, as it is, looks like:
#!/bin/bash # FILE: # scanit

check_data_files() { if [ -r netwatch -a -r netnasties ] then return 0 else return 1 fi } # Main Program if check_data_files then echo "Datafiles found" else echo "One of the datafiles missing - exiting" exit 1 fi for checkuser in $* do while read buffer do while read checksite do user=echo $buffer | cut -d" " -f1 site=echo $buffer | cut -d" " -f2 if [ "$user" = "$checkuser" -a "$site" = "$checksite" ] then echo "$user visited the prohibited site $site" fi done < netnasties done < netwatch done

At the moment, we simply print out the user and site combination - no count provided. To be really effective, we should parse the file containing the
David Jones (20.01.00) Page 224

85321, Systems Administration

Chapter 8: Shell Programming

user/site combinations (netwatch), register and user/prohibited site combinations and then when we have all the combinations and count per combination, produce a report. Given our datafile checking function, the pseudo code might look like:
if data_files_exist ... else exit 1 fi check_netwatch_file produce_report exit

It might also be an idea to build in a "default" - if no username(s) are given on the command line, we go and get all the users from the /etc/passwd file:
f [ $1 ] then the_user_list=$* else get_passwd_users fi

Exercise
8.14.

Write the shell function get_passwd_users. This function goes through the /etc/passwd file and creates a list of usernames. (Hint: username is field one of the password file, the delimiter is ":")

eval the wonderful!

The use of eval is perhaps one of the more difficult concepts in shell programming to grasp is the use of eval. eval effectively says parse (or execute) the following twice. What this means is that any shell variables that appear in the string are substituted with their real value on the first parse, then used as-they-are for the second parse. The use of this is difficult to explain without an example, so well refer back to our case study problem. The real challenge to this program is how to actually store a count of the user and site combination. The following is how I'd do it:
checkfile() { # Goes through the netwatch file and saves user/site # combinations involving sites that are in the "restricted" # list while read buffer do username=echo $buffer | cut -d" " -f1 # Get the username # Remove .s from the string site=`echo $buffer | cut -d" " -f2 | sed s/\\\.//g` for checksite in $badsites do checksite=`echo $checksite | sed s/\\\.//g`

David Jones (20.01.00)

Page 225

85321, Systems Administration

Chapter 8: Shell Programming

# Do this for the compare sites if [ "$site" = "$checksite" ] then usersite="$username$checksite" # Does the VARIABLE called $usersite exist? Note use of eval if eval [ \$$usersite ] then eval $usersite=\expr \$$usersite + 1\ else eval $usersite=1 fi fi done done < netwatch }

There are only two really tricky lines in this function:

1. site=echo $buffer | cut -d" " -f2 | sed s/\\\.//g

Creates a variable site; if buffer (one line of netwatch) contained

rabid.dog.com

then site would become:

rabiddogcom

The reason for this is because of the variable usersite:

usersite="$username$checksite"

What we are actually creating is a variable name, stored in the variable usersite - why (you still ask) did we remove the "."s? This becomes clearer when we examine the second tricky line:
2. eval $usersite=\expr \$$usersite + 1\

Remember eval "double" or "pre" parses a line - after eval has been run, you get a line which looks something like:
# $user="jamiesobrabiddogcom" jamiesobrabiddogcom=expr $jamiesobrabiddogcom + 1

What should become clearer is this: the function reads each line of the netwatch file. If the site in the netwatch file matches one of the sites stored in netnasties file (which has been cated into the variable badsites) then we store the user/site combination. We do this by first checking if there exists a variable by the name of the user/site combination - if one does exist, we add 1 to the value stored in the variable. If there wasnt a variable with the name of the user/site combination, then we create one by assigning it to "1". At the end of the function, we should have variables in memory for all the user/prohibited site combinations found in the netwatch file, something like:
jamiesobmucusslimecom=3 tonsloyemucusslimecom=1 tonsloyeboysfunnetcomfr=3 tonsloyewarezundergr=1 rootwarzundergr=4

David Jones (20.01.00)

Page 226

85321, Systems Administration

Chapter 8: Shell Programming

Note that this would be the case even if we were only interested in the users root and jamiesob. So why didnt we check in the function if the user in the netwatch file was one of the users we were interested in? Why should we!? All that does is adds an extra loop:
for every line in the file for every site for every user do check create variable if user and if site in userlist, badsitelist

whereas all we have now is

for every line in the file for every site create variable if site in badsitelist

We are still going to have to go through every user/badsite combination anyway when we produce the report - why add the extra complexity? You might also note that there is minimal file IO - datafiles are only read ONCE - lists (memory structures) may be read more than once. Exercise
8.15.

Given the checksite function, write a function called produce_report that accepts a list of usernames and finds all user/badsite combinations stored by checkfile. This function should echo lines that look something like:
jamiesob: tonsloye: tonsloye: tonsloye: mucus.slime.com 3 mucus.slime.com 1 xboys.funnet.com.fr 3 warez.under.gr 1

Step-by-step
In this section, we will examine a complex shell programming problem and work our way through the solution.

The problem
This problem is an adaptation of the problem used in the 1997 shell programming assignment for systems administration: Problem Definition Your department's FTP server provides anonymous FTP access to the /pub area of the filesystem - this area contains subdirectories (given by unit code) which contain resource materials for the various subjects offered. You suspect that this service isn't being used any more with the advent of the WWW, however, before you close this service and use the file space for something more useful, you need to prove this.

David Jones (20.01.00)

Page 227

85321, Systems Administration

Chapter 8: Shell Programming

What you require is a program that will parse the FTP logfile and produce usage statistics on a given subject. This should include: Number of accesses per user Number of bytes transferred The number of machines which have used the area.

The program will probably be called from other scripts. It should accept (from the command line) the subject (given by the subject code) that it is to examine, followed by one or more commands. Valid commands will consist of: USERS - get a user and access count listing BYTES - bytes transmitted for the subject HOSTS - number of unique machines who have used the area

Background information A cut down version of the FTP log will be examined by our program - it will consist of:
remote host name file size in bytes name of file local username or, if guest, ID string given (anonymous FTP password)

For example:
aardvark.com 138.77.8.8 2345 112 /pub/85349/lectures.tar.gz /pub/81120/cpu.gif [email protected] [email protected]

The FTP logfile will be called /var/log/ftp.log - we need not concern ourselves how it is produced (for those that are interested - look at man ftpd for a description of the real log file). Anonymous FTP usernames are recorded as whatever the user types in as the password - while this may not be accurate, it is all we have to go on. We can assume that all directories containing subject material branch off the /pub directory, eg.
/pub/85321 /pub/81120

Expected interaction The following are examples of interaction with the program (scanlog):
[david@faile david]$ scanlog 85321 USERS [email protected] 1 [email protected] 22 jonesd 56 [david@faile david]$ scanlog 85321 BYTES 2322323 [david@faile david]$ scanlog 85321 HOSTS 5

David Jones (20.01.00)

Page 228

85321, Systems Administration

Chapter 8: Shell Programming

[david@faile david]$ scanlog 85321 BYTES USERS 2322323 [email protected] 1 [email protected] 22 jonesd 56

Solving the problem

How would you solve this problem? What would you do first? Break it up What does the program have to do? What are its major parts? Lets look at the functionality again - our program must: get a user and access count listing produce a the byte count on files transmitted for the subject list the number unique machines who have used the area and how many times Read parameters from the command line, picking out the subject we are interested in go through the other parameters one by one, acting on each one, calling the appropriate function Terminate/clean up

To do this, our program must first:

So, this looks like a program containing three functions. Or is it? Look at the simple case first It is often easier to break down a problem by walking through a simple case first. Lets imagine that we want to get information about a subject - let's use the code 85321. At this stage, we really dont care what the action is. What happens? The program starts. We extract the first parameter from the command line. This is our subject. We might want to check if there is a first parameter - is it blank? Since we are only interested in this subject, we might want to go through the FTP log file and extract those entries we're interested in and keep this information in a temporary file. Our other option is to do this for every different action - this would probably be inefficient. Now, we want to go through the remaining parameters on the command line and act on each one. Maybe we should signal a error if we dont understand the action? At the end of our program, we should remove any temporary files we use.

David Jones (20.01.00)

Page 229

85321, Systems Administration

Chapter 8: Shell Programming

Pseudo Code If we were to pseudo code the above steps, wed get something like:
# Check to see if the first parameter is blank if first_parameter = "" then echo "No unit specified" exit fi # Find all the entries were interested in, place this in a TEMPFILE # Right - for every other parameter on the command line, we perform # some for ACTION in other_parameters do # Decide if it is a valid action - act on it or give a error done # Remove Temp file rm TEMPFILE

# Lets code this: if [ "$1" = "" ] then echo "No unit specified" exit 1 fi # Remove $1 from the parm line UNIT=$1 shift # Find all the entries we're interested in grep "/pub/$UNIT" $LOGFILE > $TEMPFILE # Right - for every other parameter on the command line, we perform some for ACTION in $@ do process_action "$ACTION" done # Remove Temp file rm $TEMPFILE

Ok, a few points to note: Notice the use of the variables LOGFILE and TEMPFILE? These would have to be defined somewhere above the code segment. We remove the first parameter from the command line and assign it to another variable. We do this using the shift command. We use grep to find all the entries in the original log file that refer to the subject we are interested in. We store these entries in a temporary file.

David Jones (20.01.00)

Page 230

85321, Systems Administration

Chapter 8: Shell Programming

The use of $@ in the loop to process the remaining parameters is important. Why did we use it? Why not $* ? Hint: 1 2 3 4 5 6 isn't 1 2 3 4 5 6 is it? We've invented a new function, process_action - we will use this to work out what to do with each action. Note that we are passing the function a parameter. Why are we enclosing it in quotes? Does it matter if we dont? Actually, in this case, it doesn't matter if we call the function with the parameter in quotes or not, as our parameters are expected to be single words. However, what if we allowed commands like:
find host 138.77.3.4

If we passed this string to a function (without quotes), it would be interpreted as:

$1=find $2=host $3=138.77.3.4

This wouldnt be entirely what we want - so, we enclose the string in quotes - producing:
$1=find host 138.77.3.4

As we mentioned, in this case, we have single word commands, so it doesnt matter, however, always try to look ahead for problems - ask yourself the figurative question - Is my code going to work in the rain?. Expand function process_action We have a function to work on - process_action. Again, we should pseudo code it, then implement it. Wait! We havent first thought about what we want it to do - always a good idea to think before you code! This function must take a parameter, determine if it is a valid action, then perform some action on it. It is an invalid action, then we should signal an error. Lets try the pseudo code first:
process_action() { # Now, Check what we have case Action in BYTES then do a function USERS then do a function HOSTS then do a function Something Else then echo esac }

to get bytes to get a user list to get an access count "Unknown command $theAction"

Right - now try the code:

David Jones (20.01.00)

Page 231

85321, Systems Administration

Chapter 8: Shell Programming

process_action() { # Translate to upper case theAction=echo $1 | tr [a-z] [A-Z] # Now, Check what we have case $theAction in USERS) getUserList ;; HOSTS) getAccessCount ;; BYTES) getBytes ;; *) echo "Unknown command $theAction" ;; esac }

Some comments on this code: Note that we translate the action command (for example bytes , users) into upper case. This is a nicety - it just means that well pick up every typing variation of the action. We use the case command to decide what to do with the action. We could have just as easily used a series of IF-THEN-ELSE-ELIF-FI statements - this becomes horrendous to code and read after about three conditions so case is a better option. As you will see, we've introduced calls to functions for each command this again breaks to code up into bite size pieces (excuse the pun ;) to code. This follows the top-down design style. We will now expand each function.

Expand Function getUserList Now might be a good time to revise what was required of our program - in particular, this function. We need to produce a listing of all the people who have accessed files relating to the subject of interest and how many times they've accessed files. Because we've separated out the entries of interest from the log file, we need no longer concern ourselves with the actual files and if they relate to the subject. We now are just interested in the users. Reviewing the log file format:
aardvark.com 2345 138.77.8.8 112 /pub/85349/lectures.tar.gz /pub/81120/cpu.gif [email protected] [email protected]

We see that user information is stored in the fourth field. If we pseudo code what we want to do, it would look something like:
for every_user_in the file do go_through_the_file_and_count_occurences print this out done

Expanding this a bit more, we get:

David Jones (20.01.00)

Page 232

85321, Systems Administration

Chapter 8: Shell Programming

extract_users_from_file for user in user_list do count = 0 while read log_file do if user = current_entry then count = count + 1 fi done echo user count done

Lets code this:

getUserList() { cut -f4 $TEMPFILE | sort > $TEMPFILE.users userList=uniq $TEMPFILE.users for user in $userList do { count=0 while read X do if echo $X | grep $user > /dev/null then count=expr $count + 1 fi done } < $TEMPFILE echo $user $count done rm $TEMPFILE.users }

Some points about this code: The first cut extracts a user list and places it in a temp file. A unique list of users is then created and placed into a variable. For every user in the list, the file is read through and each line searched for the user string. We pipe the output into /dev/null. If a match is made, count is incremented. Finally the user/count combination is printed. The temporary file is deleted.

Unfortunately, this code totally sucks. Why? There are several things wrong with the code, but the most outstanding problem is the massive and useless looping being performed - the while loop reads through the file for every user. This is bad. While loops within shell scripts are very time consuming and inefficient - they are generally avoided if, as in this case, they can be. More importantly, this script doesn't make use of UNIX commands which could simplify (and speed up!) our code. Remember: dont re-invent the wheel - use existing utilities where possible.
David Jones (20.01.00) Page 233

85321, Systems Administration

Chapter 8: Shell Programming

Lets try it again, this time without the while loop:

getUserList() { cut -f4 $TEMPFILE | sort > $TEMPFILE.users # Get user list userList=uniq $TEMPFILE.users for user in $userList # for every user... do count=grep $user $TEMPFILE.users | wc -l # count how many times they are echo $user $count # in the file done rm $TEMPFILE.users }

Much better! We've replaced the while loop with a simple grep command however, there are still problems: We dont need the temporary file Can we wipe out a few more steps? Next cut:
getUserList() { userList=cut -f4 $TEMPFILE | sort | uniq for user in $userList do echo $user grep $user $TEMPFILE | wc -l done }

Beautiful! Or is it. What about:

echo cut-f4 $TEMPFILE | sort | uniq -c

This does the same thing...or does it? If we didn't care what our output looked like, then this'd be ok - find out what's wrong with this code by trying it and the previous segment - compare the results. Hint: uniq -c produces a count of every sequential occurrence of an item in a list. What would happen if we removed the sort? How could we fix our output problem? Expand Function getAccessCount This function requires a the total number of unique hosts which have accessed the files. Again, as we've already separated out the entries of interest into a temporary file, we can just concentrate on the hosts field (field number one). If we were to pseudo code this:
create_unique_host list count = 0 for host in host_list do count = count + 1 done echo count

David Jones (20.01.00)

Page 234

85321, Systems Administration

Chapter 8: Shell Programming

From the previous function, we can see that a direct translation from pseudo code to shell isnt always efficient. Could we skip a few steps and try the efficient code first? Remember - we should try to use existing UNIX commands. How do we create a unique list? The hint is in the word unique - the uniq command is useful in extracting unique listings. What are we going to use as the input to the uniq command? We want a list of all hosts that accessed the files - the host is stored in the first field of every line in the file. Next hint - when we see the word field we can immediately assume we're going to use the cut command. Do we have to give cut any parameters? In this case, no. cut assumes (by default) that fields are separated by tabs - in our case, this is true. However, if the delimiter was anything else, we'd have to use a -d switch, followed by the delimiter. Next step - what about the output from uniq? Where does this go? We said that we wanted a count of the unique hosts - another hint - counting usually means using the wc command. The wc command (or word count command) counts characters, words and lines. If the output from the uniq command was one host per line, then a count of the lines would reveal the number of unique hosts. So what do we have?
cut f1 uniq wc -l

Right - how do we get input and save output for each command? A first cut approach might be:
cat $TEMPFILE | cut -f1 > $TEMPFILE.cut cat $TEMPFILE.cut | uniq > $TEMPFILE.uniq COUNT=`cat $TEMPFILE.uniq | wc -l` echo $COUNT

This is very inefficient; there are several reasons for this: We cat a file THREE times to get the count. We dont even have to use cat if we really try. We use temp files to store results - we could use a shell variable (as in the second last line) but is there any need for this? Remember, file IO is much slower than assignments to variables, which, depending on the situation, is slower again that using pipes. There are four lines of code - this can be completed in one!

So, removing these problems, we are left with:

getAccessCount() { echo cut -f1 $TEMPFILE | uniq | wc -l }

How does this work? The shell executes what's between and this is outputted by echo.

David Jones (20.01.00)

Page 235

85321, Systems Administration

Chapter 8: Shell Programming

This command starts with the cut command - a common misconception is that cut requires input to be piped into it - however, cut works just as well by accepting the name of a file to work with. The output from cut (a list of hosts) is piped into uniq. uniq then removes all duplicate host from the list - this is piped into wc. wc then counts the number of lines - the output is displayed.

Expand Function getBytes The final function we have to write (Yes! We are nearly finished) counts the total byte count of the files that have been accessed. This is actually a fairly simple thing to do, but as youll see, using shell scripting to do this can be very inefficient. First, some pseudo code:
total = 0 while read line from file do extract the byte field add this to the total done echo total

In shell, this looks something like:

getBytes() { bytes=0 while read X do bytefield=echo $X | cut -f2 bytes=expr $bytes + $bytefield done < $TEMPFILE echo $bytes }

...which is very inefficient (remember: looping is bad!). In this case, every iteration of the loop causes three new processes to be created, two for the first line, one for the second - creating processes takes time! The following is a bit better:
getBytes() { list=cut -f2 $TEMPFILE bytes=0 for number in $list do bytes=expr $bytes + $number done echo $bytes }

The above segment of code still has looping, but is more efficient with the use of a list of values which must be added up. However, we can get smarter:

David Jones (20.01.00)

Page 236

85321, Systems Administration

Chapter 8: Shell Programming

getBytes() { numstr=cut -f2 $TEMPFILE | sed "s/$/ + /g" expr $numstr 0 }

Do you see what weve done? The cut operation produces a list of numbers, one per line. When this is piped into sed, the end-of-line is substituted with + - note the spaces. This is then combined into a single line string and stored in the variable numstr. We then get the expr of this string - why do we put the 0 on the end? Two reasons: After the sed operation, there is an extra + on the end - for example, if the input was: 2 3 4 The output would be: 2 + 3 + 4 + This, when placed in a shell variable, looks like: 2 + 3 + 4 + ...which when evaluated, gives an error. Thus, placing a 0 at the end of the string matches the final + sign, and expr is happy What if there wasnt a byte count? What if there were no entries - expr without parameters doesn't work - expr with 0 does. So, is this the most efficient code? Within the shell, yes. Probably the most efficient code would be a call to awk and the use of some awk scripting, however that is beyond the scope of this chapter and should be examined as a personal exercise. A final note about the variables Throughout this exercise, we've referred to $TEMPFILE and $LOGFILE. These variables should be set at the top of the shell script. LOGFILE refers to the location of the FTP log. TEMPFILE is the actual file used to store the entries of interest. This must be a unique file and should be deleted at the end of the script. It'd be an excellent idea to store this file in the /tmp directory (just in case your script dies and you leave the temp file laying around - /tmp is regularly cleaned out by the system) - it would be an even better idea to

David Jones (20.01.00)

Page 237

85321, Systems Administration

Chapter 8: Shell Programming

guarantee its uniqueness by including the process ID ($$) somewhere within its name:
LOGFILE="/var/log/ftp.log" TEMPFILE="/tmp/scanlog.$$"

The final program - a listing

The following is the completed shell script - notice how short the code is (think of what it would be like if we hadnt been pushing for efficiency!).
#!/bin/sh # # FILE: # PURPOSE: # AUTHOR: # HISTORY: # # To do : #

scanlog Scan FTP log Bruce Jamieson DEC 1997

Created

Truly astounding things. Apart from that, process a FTP log and produce stats

#-------------------------# globals LOGFILE="ftp.log" TEMPFILE="/tmp/scanlog.$$" # functions

#---------------------------------------# getAccessCount # - display number of unique machines that have accessed the page getAccessCount() { echo cut -f1 $TEMPFILE }

| uniq | wc -l

#------------------------------------------------------# getUserList # - display the list of users who have acessed this page getUserList() { userList=cut -f4 $TEMPFILE | sort | uniq for user in $userList do echo $user grep $user $TEMPFILE | wc -l done } #------------------------------------------------------# getBytes # - calculate the amount of bytes transferred getBytes() {

David Jones (20.01.00)

Page 238

85321, Systems Administration

Chapter 8: Shell Programming

numstr=cut -f2 $TEMPFILE | sed "s/$/ + /g" expr $numstr 0 }

#-----------------------------------------------------# process_action # Based on the passed string, calls one of three functions # process_action() { # Translate to upper case theAction=echo $1 | tr [a-z] [A-Z] # Now, Check what we have case $theAction in BYTES) getBytes ;; USERS) getUserList ;; HOSTS) getAccessCount ;; *) echo "Unknown command $theAction" ;; esac }

#---#

Main

if [ "$1" = "" ] then echo "No unit specified" exit 1 fi UNIT=$1 # Remove $1 from the parm line shift # Find all the entries were interested in grep "/pub/$UNIT" $LOGFILE > $TEMPFILE # Right - for every parameter on the command line, we perform some for ACTION in $@ do process_action "$ACTION" done # Remove Temp file rm $TEMPFILE # Were finished!

Final notes
Throughout this chapter we have examined shell programming concepts including:

David Jones (20.01.00)

Page 239

85321, Systems Administration

Chapter 8: Shell Programming

variables comments condition statements repeated action commands functions recursion traps efficiency, and structure

Be aware that different shells support different syntax - this chapter has dealt with bourne shell programming only. As a final issue, you should at some time examine the Perl programming language as it offers the full functionality of shell programming but with added, compiled-code like features - it is often useful in some of the more complex system administration tasks.

Review Questions
8.1 Write a function that equates the username in the scanit program with the users full name and contact details from the /etc/passwd file. Modify scanit so its output looks something like:
*** Restricted Site Report *** The following is a list of prohibited sites, users who have visited them and on how many occasions Bruce Jamieson Elvira Tonsloy Elvira Tonsloy Elvira Tonsloy x9999 x1111 x1111 x1111 mucus.slime.com 3 mucus.slime.com 1 xboys.funnet.com.fr 3 warez.under.gr 1

(Hint: the fifth field of the passwd file usually contains the full name and phone extension (sometimes)) 8.2 Modify scanit so it produces a count of unique user/badsite combinations like the following:
*** Restricted Site Report *** The following is a list of prohibited sites, users who have visited them and on how many occasions Bruce Jamieson x9999 mucus.slime.com 3 Elvira Tonsloy x1111 mucus.slime.com 1

David Jones (20.01.00)

Page 240

85321, Systems Administration

Chapter 8: Shell Programming

Elvira Tonsloy x1111 xboys.funnet.com.fr 3 Elvira Tonsloy x1111 warez.under.gr 1 4 User/Site combinations detected.

8.3 Modify scanit so it produces a message something like:

There were no users found accessing prohibited sites! if there were no user/badsite combinations.

References
Kochan S.G. et al Jones, D "UNIX Shell Programming" SAMS 1993, USA "Shell Programming" WWW Notes

Newmarch, J "Shell Programming" http://pandonia.canberra.edu.au/OS/13_1.html

Source of scanit
#!/bin/bash # # AUTHOR: # DATE: # PROGRAM: # PURPOSE: # # # # # FILES: # # # # NOTES: # # # # # HISTORY: # Bruce Jamieson Feb 1997 scanit Program to analyse the output from a network monitor. "scanit" accepts a list of users to and a list of "restricted" sites to compare with the output from the network monitor. scanit netwatch netnasties shell script output from network monitor restricted site file

This is a totally made up example - the names of persons or sites used in data files are not in anyway related to reality - any similarity is purely coincidental :) bleak and troubled :)

checkfile() { # Goes through the netwatch file and saves user/site # combinations involving sites that are in the "restricted" # list while read buffer do username=echo $buffer | cut -d" " -f1 site=echo $buffer | cut -d" " -f2 | sed s/\\\.//g for checksite in $badsites do checksite=echo $checksite | sed s/\\\.//g

David Jones (20.01.00)

Page 241

85321, Systems Administration

Chapter 8: Shell Programming

# echo $checksite $site if [ "$site" = "$checksite" ] then usersite="$username$checksite" if eval [ \$$usersite ] then eval $usersite=\expr \$$usersite + 1\ else eval $usersite=1 fi fi done done < netwatch } produce_report() { # Goes through all possible combinations of users and # restricted sites - if a variable exists with the combination, # it is reported for user in $* do for checksite in $badsites do writesite=echo $checksite checksite=echo $checksite | sed s/\\\.//g usersite="$user$checksite" if eval [ \$$usersite ] then eval echo "$user: $writesite \$$usersite" usercount=expr $usercount + 1 fi done done } get_passwd_users() { # Creates a user list based on the /etc/passwd file while read buffer do username=echo $buffer | cut -d":" -f1 the_user_list=echo $username $the_user_list done < /etc/passwd } check_data_files() { if [ -r netwatch -a -r netnasties ] then return 0 else return 1 fi } # Main Program # Uncomment the next line for debug mode #set -x

David Jones (20.01.00)

Page 242

85321, Systems Administration

Chapter 8: Shell Programming

if check_data_files then echo "Datafiles found" else echo "One of the datafiles missing - exiting" exit 1 fi usercount=0 badsites=cat netnasties if [ $1 ] then the_user_list=$* else get_passwd_users fi echo echo "*** Restricted Site Report ***" echo echo The following is a list of prohibited sites, users who have echo visited them and on how many occasions echo checkfile produce_report $the_user_list echo if [ $usercount -eq 0 ] then echo "There were no users found accessing prohibited sites!" else echo "$usercount prohibited user/site combinations found." fi echo echo # END scanit

David Jones (20.01.00)

Page 243

85321, Systems Administration

Chapter 9: Users

Chapter
Users

Introduction
Before anyone can use your system they must have an account. This chapter examines user accounts and the responsibilities of the Systems Administrators with regards to accounts. By the end of this chapter you should be aware of the process involved in creating and removing user accounts, be familiar with the configuration files that UNIX uses to store information about accounts, know what information you must have to create an account, understand the implications of choosing particular usernames, user ids and passwords, be aware of special accounts including the root account and the implications of using the root account, have been introduced to a number of public domain tools that help with account management.

Other Resources
Other material which discusses user and authentication related material includes Guides The Linux Installation and Getting Started Guide has a section (4.6) on user management. The Linux Systems Administrators Guides chapter 9 also discusses managing user accounts. The Linux Administration Made Easy Guide also provides some discussion of account related issues in its chapter 6, General System Administration Issues. The RedHat 6.0 Guides also mention account management issues.

What is a UNIX account?

A UNIX account is a collection of logical characteristics that specify who the user is, what the user is allowed to do and where the user is allowed to do it. These characteristics include a login (or user) name, password, numeric user identifier or UID,

David Jones (20.01.00)

Page 244

85321, Systems Administration

Chapter 9: Users

a default numeric group identifier or GID, Many accounts belong to more than one group but all accounts have one default group. home directory, login shell, possibly a mail alias, mail file, and collection of startup files.

Login names
The account of every user is assigned a unique login (or user) name. The username uniquely identifies the account for people. The operating system uses the user identifier number (UID) to uniquely identify an account. The translation between UID and the username is carried out reading the /etc/passwd file (/etc/passwd is introduced below). Login name format On a small system, the format of login names is generally not a problem since with a small user population it is unlikely that there will be duplicates. However on a large site with hundreds or thousands of users and multiple computers, assigning a login name can be a major problem. With a larger number of users it is likely that you may get a number of people with names like David Jones, Darren Jones. The following is a set of guidelines. They are by no means hard and fast rules but using some or all of them can make life easier for yourself as the Systems Administrator, or for your users. unique This means usernames should be unique not only on the local machine but also across different machines at the same site. A login name should identify the same person and only one person on every machine on the site. This can be very hard to achieve at a site with a large user population especially if different machines have different administrators. The reason for this guideline is that under certain circumstances it is possible for people with the same username to access accounts with the same username on different machines. There is an increasing trend for global logons. One username/password will get users into all of the systems they need for a given organisation. Not quite there but getting there. up to 8 characters UNIX will ignore or disallow login names that are longer. Dependent on which platform you are using. Lowercase Numbers and upper case letters can be used. Login names that are all upper

David Jones (20.01.00)

Page 245

85321, Systems Administration

Chapter 9: Users

case should be avoided as some versions of UNIX can assume this to mean your terminal doesnt recognise lower case letters and every piece of text subsequently sent to your display is in uppercase. Easy to remember A random sequence of letters and numbers is hard to remember and so the user will be continually have to ask the Systems Administrator "whats my username?" No nicknames A username will probably be part of an email address. The username will be one method by which other users identify who is on the system. Not all the users may know the nicknames of certain individuals. A fixed format There should be a specified system for creating a username. Some combination of first name, last name and initials is usually the best. Setting a policy allows you to automate the procedure of adding new users. It also makes it easy for other users to work out what the username for a person might be.

Passwords
An accounts password is the key that lets someone in to use the account. A password should be a secret collection of characters known only by the owner of the account. Poor choice of passwords is the single biggest security hole on any multi-user computer system. As a Systems Administrator you should follow a strict set of guidelines for passwords (after all if someone can break the root accounts password, your system is going bye, bye). In addition you should promote the use of these guidelines amongst your users. Password guidelines An example set of password guidelines might include use combinations of upper and lower case characters, numbers and punctuation characters, dont use random combinations of characters if they break the following two rules, be easy to remember, If a user forgets their password they cant use the system and guess whom they come and see. Also the user SHOULD NOT have to write their password down. be quick to type, One of the easiest and most used methods for breaking into a system is simply watching someone type in their password. It is harder to do if the password is typed in quickly.

David Jones (20.01.00)

Page 246

85321, Systems Administration

Chapter 9: Users

a password should be at least 6 characters long, The shorter a password is the easier it is to break. Some systems will not allow passwords shorter than a specified length. a password should not be any longer than 8 to 10 characters, Most systems will look as if they are accepting longer passwords but they simply ignore the extra characters. The actual size is system specific but between eight and ten characters is generally the limit. do not use words from ANY language, Passwords that are words can be cracked (youll see how later). do not use combinations of just words and numbers, Passwords like hello1 are just as easy to crack as hello. use combinations of words separated by punctuation characters or acronyms of uncommon phrases/song lines, They should be easy to remember but hard to crack. e.g. b1gsh1p change passwords regularly, Not too often that you forget which password is currently set. never reuse passwords.

The UID
Every account on a UNIX system has a unique user or login name that is used by users to identify that account. The operating system does not use this name to identify the account. Instead each account must be assigned a unique user identifier number (UID) when it is created. The UID is used by the operating system to identify the account. UID guidelines In choosing a UID for a new user there are a number of considerations to take into account including choose a UID number between 100 and 32767 (or 60000), Numbers between 0 and 99 are reserved by some systems for use by system accounts. Different systems will have different possible maximum values for UID numbers. Around 32000 and 64000 are common upper limits. UIDs for a user should be the same across machines, Some network systems (e.g. NFS) require that users have the same UID across all machines in the network. Otherwise they will not work properly. you may not want to reuse a number. Not a hard and fast rule. Every file is owned by a particular user id. Problems arise where a user has left and a new user has been assigned the UID of the old user. What happens when you restore from backups some files that were created by the old user? The file thinks the user with a particular UID owns it. The new user will now own those files even though the username has changed.

David Jones (20.01.00)

Page 247

85321, Systems Administration

Chapter 9: Users

Home directories
Every user must be assigned a home directory. When the user logs in it is this home directory that becomes the current directory. Typically all user home directories are stored under the one directory. Many modern systems use the directory /home. Older versions used /usr/users. The names of home directories will match the username for the account. For example, a user jonesd would have the home directory /home/jonesd In some instances it might be decided to further divide users by placing users from different categories into different sub-directories. For example, all staff accounts may go under /home/staff while students are placed under /home/students. These separate directories may even be on separate partitions.

Login shell
Every user account has a login shell. A login shell is simply the program that is executed every time the user logs in. Normally it is one of the standard user shells such as Bourne, csh, bash etc. However it can be any executable program. One common method used to disable an account is to change the login shell to the program /bin/false. When someone logs into such an account /bin/false is executed and the login: prompt reappears.

Dot files
A number of commands, including vi, the mail system and a variety of shells, can be customised using dot files. A dot file is usually placed into a users home directory and has a filename that starts with a . (dot). This files are examined when the command is first executed and modifies how it behaves. Dot files are also known as rc files. As you shouldve found out by doing one of the exercises from the previous chapter rc is short for "run command" and is a left over from an earlier operating system. Commands and their dot files Table 9.1 summarises the dot files for a number of commands. The FAQs for the newsgroup comp.unix.questions has others. Shell dot files These shell dot files, particularly those executed when a shell is first executed, are responsible for setting up command aliases, Some shells (e.g. bash) allow the creation of aliases for various commands. A common command alias for old MS-DOS people is dir, usually set to mean the same as ls -l. setting values for shell variables like PATH and TERM.
Page 248

David Jones (20.01.00)

85321, Systems Administration

Chapter 9: Users

Filename ~/.cshrc ~/.login

Command /bin/csh /bin/csh

Explanation Executed every time C shell started. Executed after .cshrc when logging in with C shell as the login shell. Executed during the login of every user that uses the Bourne shell or its derivatives. Located in users home directory. Executed whenever the user logs in when the Bourne shell is their login shell executed just prior to the system logging the user out (when the csh is the login shell) executed just prior to the system logging the user out (when bash is the login shell) records the list of commands executed using the current shell Used to forward mail to another address or a command used to set options for use in vi
Table 9.1 Dot files

/etc/profile

/bin/sh

~/.profile

/bin/sh

~/.logout

/bin/csh

~/.bash_logout

/bin/bash

~/.bash_history

/bin/bash

~/.forward

incoming mail
vi

~/.exrc

Skeleton directories
Normally all new users are given the same startup files. Rather than create the same files from scratch all the time, copies are usually kept in a directory called a skeleton directory. This means when you create a new account you can simply copy the startup files from the skeleton directory into the users home directory. The standard skeleton directory is /etc/skel. It should be remembered that the files in the skeleton directory are dot files and will not show up if you simply use ls /etc/skel. You will have to use the -a switch for ls to see dot files. Exercises
9.1.

Examine the contents of the skeleton directory on your system (if you have one). Write a command to copy the contents of that directory to another. Hint: Its harder than it looks. Use the bash dot files to create an alias dir that performs the command
ls -al

9.2.

David Jones (20.01.00)

Page 249

85321, Systems Administration

Chapter 9: Users

The mail file

When someone sends mail to a user that mail message has to be stored somewhere so that it can be read. Under UNIX each user is assigned a mail file. All user mail files are placed in the same directory. When a new mail message arrives it is appended onto the end of the users mail file. The location of this directory can change depending on the operating system being used. Common locations are
/usr/spool/mail, /var/spool/mail,

This is the standard Linux location.

/usr/mail /var/mail.

All mail in the one location On some sites it is common for users to have accounts on a number of different computers. It is easier if all the mail for a particular user goes to the one location. This means that a user will choose one machine as their mail machine and want all their email forwarded to their account on that machine. There are at least two ways by which mail can be forwarded the user can create a .forward file in their home directory (see Table 7.1), or the administrator can create an alias.

Mail aliases
If you send an e-mail message that cannot be delivered (e.g. you use the wrong address) typically the mail message will be forwarded to the postmaster of your machine. There is usually no account called postmaster, postmaster is a mail alias. When the mail delivery program gets mail for postmaster it will not be able to find a matching username. Instead it will look up a specific file, under Linux /etc/aliases. This file will typically have an entry like
postmaster: root

This tells the delivery program that anything addressed postmaster should actually be delivered to the user root. Take a look at the /etc/aliases file on your system for other aliases. Site aliases Some companies will have a set policy for e-mail aliases for all staff. This means that when you add a new user you also have to update the aliases file.

David Jones (20.01.00)

Page 250

85321, Systems Administration

Chapter 9: Users

For example The Central Queensland University has aliases set up for all staff. An e-mail with an address using the format [email protected] will be delivered to that staff members real mail address. In my case the alias is [email protected]. The main on-campus mail host has an aliases file that translates this alias into my actual e-mail address [email protected]. Linux mail The following exercise requires that you have mail delivery working on your system. You can test whether or not email is working on your system by starting one of the provided email programs (e.g. elm) and send yourself an email message. You do this by using only your username as the address (no @). If it isnt working, refer to the documentation from RedHat on how to get email functioning. Exercises
9.3.

Send a mail message from the root user to your normal user account using a mail program of your choice. Send a mail message from the root user to the address notHere. This mail message should bounce (be unable to be delivered). You will get a returned mail message. Have a look at the mail file for postmaster. Has it increased? Create an alias for notHere and try the above exercise again. If you have installed sendmail, the following steps should create an alias - login as root, - add a new line containing notHere: root in the file /etc/aliases - run the command newaliases

9.4.

9.5.

Account configuration files

Most of the characteristics of an account mentioned above are stored in two or three configuration files. All these files are text files, each account has a oneline entry in the file with each line divided into a number of fields using colons. Table 9.2. lists the configuration files examined and their purpose. Not all systems will have the /etc/shadow file. By default Linux doesnt however it is possible to install the shadow password system. On some platforms the shadow file will exist but its filename will be different.

David Jones (20.01.00)

Page 251

85321, Systems Administration

Chapter 9: Users

File
/etc/passwd

Purpose the password file, holds most of an account characteristics including username, UID, GID, GCOS information, login shell, home directory and in some cases the password the shadow password file, a more secure mechanism for holding the password, common on more modern systems the group file, holds characteristics about a systems groups including group name, GID and group members
Table 9.2 Account configuration files

/etc/shadow

/etc/group

/etc/passwd /etc/passwd is the main account configuration file. Table 9.3 summarises each of the fields in the /etc/passwd file. On some systems the encrypted password will not be in the passwd file but will be in a shadow file. Field Name Purpose

login name

the users login name

encrypted password * encrypted version of the users password UID number default GID GCOS information the users unique numeric identifier the users default group id no strict purpose, usually contains full name and address details, sometimes called the comment field the directory in which the user is placed when they log in the program that is run when the user logs in
shadow 9.3 password file Table

home directory login shell

* not

on system s with a

/etc/passwd

Exercises
9.6.

Examine your accounts entry in the /etc/passwd field. What is your UID, GID? Where is your home directory and what is your login shell?

David Jones (20.01.00)

Page 252

85321, Systems Administration

Chapter 9: Users

Everyone can read /etc/passwd

Every user on the system must be able to read the /etc/passwd file. This is because many of the programs and commands a user executes must access the information in the file. For example, when you execute the command ls -l command part of what the command must do is translate the UID of the files owner into a username. The only place that information is stored is in the /etc/passwd file.

This is a problem
Since everyone can read the /etc/passwd file they can also read the encrypted password. The problem isnt that someone might be able to decrypt the password. The method used to encrypt the passwords is supposedly a one way encryption algorithm. You arent supposed to be able to decrypt the passwords.

Password matching
The way to break into a UNIX system is to obtain a dictionary of words and encrypt the whole dictionary. You then compare the encrypted words from the dictionary with the encrypted passwords. If you find a match you know what the password is. Studies have shown that with a carefully chosen dictionary, between 10-20% of passwords can be cracked on any machine. Later in this chapter youll be shown a program that can be used by the Systems Administrator to test users passwords. An even greater problem is the increasing speed of computers. One modern super computer is capable of performing 424,400 encryptions a second. This means that all six-character passwords can be discovered in two days and all seven-character passwords within four months.

The solution
The solution to this problem is to not store the encrypted password in the /etc/passwd file. Instead it should be kept in another file that only the root user can read. Remember the passwd program is setuid root. This other file in which the password is stored is usually referred to as the shadow password file. It can be stored in one of a number of different locations depending on the version of UNIX you are using. A common location, and the one used by the Linux shadow password suite, is /etc/shadow. During installation of Redhat 6.1 you are given the choice of using shadow passwords or not. Where possible you should use shadow passwords.

David Jones (20.01.00)

Page 253

85321, Systems Administration

Chapter 9: Users

Shadow file format

Typically the shadow file consists of one line per user containing the encrypted password and some additional information including username, the date the password was last changed, minimum number of days before the password can be changed again, maximum number of days before the password must be changed, number of days until age warning is sent to user, number of days of inactivity before account should be removed, absolute date on which the password will expire.

The additional information is used to implement password aging. This will be discussed later in the security chapter.

Groups
A group is a logical collection of users. Users with similar needs or characteristics are usually placed into groups. A group is a collection of user accounts that can be given special permissions. Groups are often used to restrict access to certain files and programs to everyone but those within a certain collection of users.
/etc/group

The /etc/group file maintains a list of the current groups for the system and the users that belong to each group. The fields in the /etc/group file include the group name, A unique name for the group. an encrypted password (this is rarely used today) , the numeric group identifier or GID, and the list of usernames of the group members separated by commas.

For example On the Central Queensland University UNIX machine jasper only certain users are allowed to have full Internet access. All these users belong to the group called angels. Any program that provides Internet access has as the group owner the group angels and is owned by root. Only members of the angels group or the root user can execute these files.

David Jones (20.01.00)

Page 254

85321, Systems Administration

Chapter 9: Users

The default group Every user is the member of at least one group sometimes referred to as the default group. The default group is specified by the GID specified in the users entry in the /etc/passwd file. Since the default group is specified in /etc/passwd it is not necessary for the username to be added to the /etc/group file for the default group. Other groups A user can in fact be a member of several groups. Any extra groups the user is a member of are specified by entries in the /etc/group file. It is not necessary to have an entry in the /etc/group file for the default group. However if the user belongs to any other groups they must be added to the /etc/group file.

Special accounts
All UNIX systems come with a number of special accounts. These accounts already exist and are there for a specific purpose. Typically these accounts will all have UIDs that are less than 100 and are used to perform a variety of administrative duties. Table 9.4. lists some of the special accounts that may exist on a machine. Purpose Username UID
root 0

The super user account. Used by the Systems Administrator to perform a number of tasks. Can do anything. Not subject to any restrictions Owner of many of the system daemons (programs that run in the background waiting for things to happen). The owner of many of the standard executable programs

daemon

bin

Table 9.4 Special accounts

root
The root user, also known as the super user is probably the most important account on a UNIX system. This account is not subject to the normal restrictions placed on standard accounts. It is used by the Systems

David Jones (20.01.00)

Page 255

85321, Systems Administration

Chapter 9: Users

Administrator to perform administrative tasks that cant be performed by a normal account.

Restricted actions
Some of the actions for which youd use the root account include creating and modifying user accounts, shutting the system down, configuring hardware devices like network interfaces and printers, changing the ownership of files, setting and changing quotas and priorities, and setting the name of a machine.

Be careful
You should always be careful when logged in as root. When logged in as root you must know what every command you type is going to do. Remember the root account is not subject to the normal restrictions of other accounts. If you execute a command as root it will be done, whether it deletes all the files on your system or not.

The mechanics
Adding a user is a fairly mechanical task that is usually automated either through shell scripts or on many modern systems with a GUI based program. However it is still important that the Systems Administrator be aware of the steps involved in creating a new account. If you know how it works you can fix any problems which occur. The steps to create a user include adding an entry for the new user to the /etc/passwd file, setting an initial password, adding an entry to the /etc/group file, creating the users home directory, creating the users mail file or setting a mail alias, creating any startup files required for the user, testing that the addition has worked, and possibly sending an introductory message to the user.

Other considerations
This chapter talks about account management which includes the mechanics of adding a new account. User management is something entirely different. When adding a new account, user management tasks that are required include
David Jones (20.01.00) Page 256

85321, Systems Administration

Chapter 9: Users

making the user aware of the sites policies regarding computer use, getting the user to sign an "acceptable use" form, letting the user know where and how they can find information about their system, and possibly showing the user how to work the system.

Pre-requisite Information
Before creating a new user there is a range of information that you must know including the username format being used at your site, Are you using djones jonesdd david jones or perhaps youre using student or employee numbers for usernames. the users name and other person information, Phone number, are they a computing person, someone from sales? where the users home directory will be, will this user need a mail file on this machine or should there be an alias set up, startup shell, startup files, UID and GID. Again there should be some site wide standard for this.

Adding an /etc/passwd entry

For every new user, an entry has to be added to the /etc/passwd file. There are a variety of methods by which this is accomplished including using an editor, This is a method that is often used. However it can be unsafe and it is generally not a good idea to use it. the command vipw, or Some systems (usually BSD based) provide this command. vipw invokes an editor so the Systems Administrator can edit the passwd file safely. The command performs some additional steps that ensures that the editing is performed consistently. Some distributions of Linux supply vipw. a dedicated adduser program. Many systems, Linux included, provide a program (the name will change from system to system) that accepts a number of command-line parameters and then proceeds to perform many of the steps involved in creating a new account. The Linux command is called adduser

David Jones (20.01.00)

Page 257

85321, Systems Administration

Chapter 9: Users

The initial password

NEVER LEAVE THE PASSWORD FIELD BLANK. If you are not going to set a password for a user put a * in the password field of /etc/passwd or the /etc/shadow file. On most systems, the * character is considered an invalid password and it prevents anyone from using that account. If a password is to be set for the account then the passwd command must be used. The user should be forced to immediately change any password set by the Systems Administrator

/etc/group entry
While not strictly necessary, the /etc/group file should be modified to include the users login name in their default group. Also if the user is to be a member of any other group they must have an entry in the /etc/group file. Editing the /etc/group file with an editor should be safe.

The home directory

Not only must the home directory be created but the permissions also have to be set correctly so that the user can access the directory. The permissions of a home directory should be set such that the user should be the owner of the home directory, the group owner of the directory should be the default group that the user belongs to, at the very least, the owner of the directory should have rwx permissions, and the group and other permissions should be set as restrictively as possible.

The startup files

Once the home directory is created the startup files can be copied in or created. Again you should remember that this will be done as the root user and so root will own the files. You must remember to change the ownership. For example The following is an example set of commands that will perform these tasks.
mkdir home_directory cp -pr /etc/skel/.[a-zA-Z]* home_directory chown -R login_name home_directory chgrp -R group_name home_directory chmod -R 700 home_directory

David Jones (20.01.00)

Page 258

85321, Systems Administration

Chapter 9: Users

Setting up mail
A new user will either want to read their mail on this machine, or want to read their mail on another machine. The users choice controls how you configure the users mail. A mail file If the user is going to read their mail on this machine then you must create them a mail file. The mail file must go in a standard directory (usually /var/spool/mail under Linux). As with home directories it is important that the ownership and the permissions of a mail file be set correctly. The requirements are the user must be able to read and write the file, After all, the user must be able to read and delete mail messages. the group owner of the mail file should be the group mail and the group should be able to read and write to the file, The programs that deliver mail are owned by the group mail. These programs must be able to write to the file to deliver the users mail. no-one else should have any access to the file, No-one wants anyone else peeking at their private mail.

Mail aliases and forwards If the users main mail account is on another machine, any mail that is sent to this machine should be forwarded to the appropriate machine. There are two methods a mail alias, or a file ~/.forward Both methods achieve the same result. The main difference is that the user can change the .forward file if they wish to. They cant modify a central alias.

Testing an account
Once the account is created, at least in some instances, you will want to test the account creation to make sure that it has worked. There are at least two methods you can use login as the user use the su command.

David Jones (20.01.00)

Page 259

85321, Systems Administration

Chapter 9: Users

The su command The su command is used to change from one user account to another. To a certain extent it acts like logging in as the other user. The standard format is su username.
[david@beldin david]$ su Password:

Time to become the root user. su without any parameter lets you become the root user, as long as you know the password. In the following the id command is used to prove that I really have become the root user. Youll also notice that the prompt displayed by the shell has changed as well. In particular notice the # character, commonly used to indicate a shell with root permission.
[root@beldin david]# id uid=0(root) gid=0(root) groups=0(root),1(bin),2(daemon),3(sys),4(adm),6(disk),10(wheel) [root@beldin david]# pwd /home/david

Another point to notice is that when you dont use the "" argument for su all that has changed is user and group ids. The current directory doesnt change.
[root@beldin david]# cd / [root@beldin /]# pwd / [root@beldin /]# su david [david@beldin /]$ pwd / [david@beldin /]$ exit

However, when you do use the "" argument of the su command, it simulates a full login. This means that any startup files are executed and that the current directory becomes the home directory of the user account you "are becoming". This is equivalent to logging in as the user.
[root@beldin /]# su david [david@beldin david]$ pwd /home/david

If you run su as a normal user you will have to enter the password of the user you are trying to become. If you dont specify a username you will become the root user (if you know the password). The "" switch The su command is used to change from one user to another. By default, su david will change your UID and GID to that of the user david (if you know the password) but wont change much else. Using the - switch of su it is possible to simulate a full login including execution of the new users startup scripts and changing to their home directory.

David Jones (20.01.00)

Page 260

85321, Systems Administration

Chapter 9: Users

su

as root

If you use the su command as the root user you do not have to enter the new users password. su will immediately change you to the new user. su especially with the - switch is useful for testing a new account. Exercises
9.7.

Login as yourself and perform the following steps - show your current directory (use the pwd command), - show you current user id and group id (use the id command), - use su to become the root user, - repeat the first two steps - use the command "su " to simulate a full login as the root user, - repeat the first two steps Whats the difference between using su and su -?

9.8.

Inform the user

Lastly you should inform the user of their account details. Included in this should be some indication of where they can get assistance and some pointers on where to find more documentation. Exercises
9.9.

By hand, create a new account for a user called David Jones.

Removing an account
Deleting an account involves reversing the steps carried out when the account was created. It is a destructive process and whenever something destructive is performed, care must always be taken. The steps that might be carried out include disabling the account, backing up and removing the associated files setting up mail forwards.

Situations under which you may wish to remove an account include as punishment for a user who has broken the rules, or In this situation you may only want to disable the account rather than remove it completely. an employee has left.

David Jones (20.01.00)

Page 261

85321, Systems Administration

Chapter 9: Users

Disabling an account
Disabling an account ensures that no-one can login but doesnt delete the contents of the account. This is a minimal requirement for removing an account. There are two methods for achieving this change the login shell, or Setting the login shell to /bin/false will prevent logins. However it may still be possible for the user to receive mail through the account using POP mail programs like Eudora. change the password.

The * character is considered by the password system to indicate an illegal password. One method for disabling an account is to insert a * character into the password field. If you want to re-enable the account (with the same password) simply remove the *. Another method is to simply remove the entry from the /etc/passwd and /etc/shadow files all together. Backing up It is possible that this user may have some files that need to be used by other people. So back everything up, just in case. Remove the users files All the files owned by the account should be removed from wherever they are in the file hierarchy. It is unlikely for a user to own files that are located outside of their home directory (except for the mail file). However it is a good idea to search for them. Another use for the find command. Mail for old users On some systems, even if you delete the users mail file, mail for that user can still accumulate on the system. If you delete an account entirely by removing it from the password field, any mail for that account will bounce. In most cases, a user who has left will want their mail forwarded onto a new account. One solution is to create a mail alias for the user that points to their new address.

The Goals of Account Creation

As mentioned previously there is little point in adding users manually. It is a simple task which can be quite easily automated. This section looks at some of the tools you can use to automate this task. There are at least three goals a Systems Administrator will want to achieve with adding users make it as simple as possible automate the addition of large numbers of users
Page 262

David Jones (20.01.00)

85321, Systems Administration

Chapter 9: Users

delegate the task of adding users to someone else

The following sections will show you the tools which will allow you to achieve these goals.

Making it simple
If youve completed exercise 9.8 you should by now be aware of what a straight forward, but time consuming, task creating a new user account is. Creating an account manually might be okay for one or two accounts but adding 100 this way would get quite annoying. Luckily there are a number of tools which make this process quite simple.

useradd
useradd is an executable program which significantly reduces the complexity of adding a new user. A solution to the previous exercise using useradd looks like this
useradd c "David Jones" david

useradd will automatically create the home directory and mail file, copy files from skeleton directories and a number of other tasks. Refer to the useradd man page for more information.

userdel and usermod

userdel is the companion command to useradd and as the name suggests it deletes or removes a user account from the system. usermod allows a Systems Administrator to modify the details of an existing user account.

Graphical Tools
RedHat Linux provides a number of tools with graphical user interfaces to help both the Systems Administrator and the normal user. Tools such as userinfo and userpasswd allow normal users to modify their user accounts. RedHat also provides a command called control-panel which provides a graphical user interface for a number of Systems Administration related tasks including user management. control-panel is in fact just a simple interface to run a number of other programs which actually perform the tasks. For example, to perform the necessary user management tasks control-panel will run the command usercfg. Diagram 9.1 provides examples of the interface provided by the usercfg command.

David Jones (20.01.00)

Page 263

85321, Systems Administration

Chapter 9: Users

Diagram 9.1 usercfg interface

Automation
Tools with a graphical user interface are nice and simple for creating one or two users. However, when you must create hundreds of user accounts, they are a pain. In situations like this you must make use of a scripting language to automate the process. The process of creating a user account can be divided into the following steps gathering the appropriate information, deciding on policy for usernames, passwords etc, creating the accounts, performing any additional special steps.

The steps in this process are fairly general purpose and could apply in any situation requiring the creation of a large number of user accounts, regardless of the operating system.

Gathering the information

The first part of this chapter described the type of information that is required in order to create a UNIX user account. When automating the large scale creation of user accounts this information is generally provided in an electronic format. Often this information will be extracted from a database and converted into the appropriate format.

Policy
Gathering the raw information is not sufficient. Policy must be developed which specifies rules such as username format, location of home directories,

David Jones (20.01.00)

Page 264

85321, Systems Administration

Chapter 9: Users

which groups users will belong to and other information discussed earlier in the chapter. There are no hard and fast rules for this policy. It is a case of applying whatever works best for your particular situation. For example CQ-PAN (http://cq-pan.cqu.edu.au) was a system managed mainly by CQU computing students. CQ-PAN provided accounts for students for a variety of reasons. During its history it has used two username formats ba format The first username format, based on that used by Freenet system, was ba005 ba103 ba321 name format This was later changed to something a little more personal, firstnameLastInitialNumber. e.g. davidj1 carolyg1

Creating the accounts

Once you know what format the user information will be in and what formats you wish to follow for user accounts, you can start creating the accounts. Generally this will use the following steps read the information from the provided data file convert it into the format specified in the site policy use the UNIX account creation commands to create the accounts

Additional steps
Simply creating the accounts using the steps introduced above is usually not all that has to be done. Most sites may include additional steps in the account creation process such as sending an initial, welcoming email message, Such an email can serve a number of purposes, including informing the new users of their rights and responsibilities. It is important that users be made aware as soon as possible of what they can and cant do and what support they can expect from the Systems Administration team. creating email aliases or other site specific steps.

Changing passwords without interaction

Quite a few years ago there was a common problem that had to be overcome in order to automate the account creation process. This problem was how to set the new users password without human intervention. Remember, when creating hundreds of accounts it is essential to remove all human interaction.

David Jones (20.01.00)

Page 265

85321, Systems Administration

Chapter 9: Users

Given that this is such a common problem for UNIX systems, there are now a number of solutions to this problem. RedHat Linux comes with a number of solutions including the commands chpasswd, newusers and mkpasswd. mkpasswd is an example of an Expect (http://expect.nist.gov/) script. Expect is a program that helps to automate interactive applications such as passwd and others including telnet ftp etc. This allows you to write scripts to automate processes which normally require human input. For example In the pre-Web days (1992), satellite weather photos were made available via FTP from a computer at James Cook University. These image files were stored using a standard filename policy which indicated which date and time the images were taken. If you wanted to view the latest weather image you had to manually ftp to the James Cook computer, download the latest image and then view it on your machine. Manually ftping the files was not a large task, only 5 or 6 separate commands, however if you were doing this five times a day it got quite repetitive. Expect provides a mechanism by which a script could be written to automate this process.

Delegation
Systems Administrators are highly paid, technical staff. A business does not want Systems Administrators wasting their time performing mundane, lowlevel, repetitive tasks. Where possible a Systems Administrator should delegate responsibility for low-level tasks to other staff. In this section we examine one approach using the sudo command.

Allocating root privilege

Many of the menial tasks, like creating users and performing backups, require the access which the root account provides. This means that these tasks cant be allocated to junior members of staff without giving them access to everything else on the system. In most cases you dont want to do this. There is another problem with the root account. If you have a number of trusted Systems Administrators the root account often becomes a group account. The problem with this is that since everyone knows the root password there is now way by which you can know who is doing what as root. There is no accountability. While this may not be a problem on your individual system on commercial systems it is essential to be able to track what everyone does.

sudo
A solution to these problems is the sudo command. sudo (http://www.courtesan.com/sudo/) is not a standard UNIX command but a widely available public domain tool. It comes standard on most Linux
David Jones (20.01.00) Page 266

85321, Systems Administration

Chapter 9: Users

distributions. It does not appear to be included with RedHat. You can find a copy of sudo on the 85321 Web site/CD-ROM under the Resource Materials section for week 5. sudo allows you to allocate certain users the ability to run programs as root without giving them access to everything. For example, you might decide that the office secretary can run the adduser script, or an operator might be allowed to execute the backup script. sudo also provides a solution to the accountability problem. sudo logs every command people perform while using it. This means that rather than using the root account as a group account, you can provide all your Systems Administrators with sudo access. When they perform their tasks with sudo, what they do will be recorded. For example To execute a command as root using sudo you login to your "normal" user account and then type sudo followed by the command you wish to execute. The following example shows what happens when you can and cant executable a particular command using sudo.
[david@mc:~]$ sudo ls We trust you have received the usual lecture from the local System Administrator. It usually boils down to these two things: #1) Respect the privacy of others. #2) Think before you type. 85321.students archive [david@mc:~]$ sudo cat Sorry, user david is not allowed to execute "/bin/cat" as root on mc.

If the sudoers file is configured to allow you to execute this command on the current machine, you will be prompted for your normal password. Youll only be asked for the password once every five minutes.
/etc/sudoers

The sudo configuration file is usually /etc/sudoers or in some instances /usr/local/etc/sudoers. sudoers is a text file with lines of the following format
username hostname=command

An example sudoers file might look like this

root david bob jo ALL=ALL ALL=ALL cq-pan=/usr/local/bin/backup ALL=/usr/local/bin/adduser

In this example the root account and the user david are allowed to execute all commands on all machines. The user bob can execute the /usr/local/bin/backup command but only on the machine cq-pan. The user jo can execute the adduser command on all machines. The sudoers man page has a more detail example and explanation.

David Jones (20.01.00)

Page 267

85321, Systems Administration

Chapter 9: Users

By allowing you to specify the names of machines you can use the same sudoers file on all machines. This makes it easier to manage a number of machines. All you do is copy the same file to all your machines (there is a utility called rdist which can make this quite simple).

sudo advantages
sudo

offers the following advantages

accountability because all commands executed using sudo are logged, Logging on a UNIX computer, as youll be shown in a later chapter, is done via the syslog system. What this means is that on a RedHat machine the information from sudo is logged in the file /var/log/messages. menial tasks can be allocated to junior staff without providing root access, using sudo is faster than using su, a list of users with root access is maintained, privileges can be revoked without changing the root password.

Some sites that use sudo keep the root password in an envelope in someones draw. The root account is never used unless in emergencies where it is required.

Exercises
9.10.

Install sudo onto your system. The source code for sudo is available from the Resource Materials section of the 83521 Website/CD-ROM. Configure your version of sudo so that you can use it as a replacement for handing out the root password. What does your /etc/sudoers file look like? Use sudo a number of times. What information is logged by the sudo command? One of the listed advantages of sudo is the ability to log what people are doing with the root access. Without some extra effort this accountability can be quite pointless. Why? (Hint: the problem only really occurs with users such as david in the above example sudoers file.

9.11.

9.12.

9.13.

Conclusions
Every user on a UNIX machine must have an account. Components of a user account can include login names (also called a username), passwords, the numeric user identifier or UID, the numeric group identifier or GID,
Page 268

David Jones (20.01.00)

85321, Systems Administration

Chapter 9: Users

a home directory, a login shell, mail aliases, a mail file, and startup files.
/etc/passwd, /etc/shadow, /etc/group, and

Configuration files related to user accounts include

to a certain extent /etc/aliases

Creating a user account is a mechanical task that can and often is automated. Creating an account also requires root privilege. Being the root user implies no restrictions and enables anything to be done. It is generally not a good idea to allocate this task to a junior member of staff. However, there are a number of tools which allow this and other tasks to be delegated.

Review Questions
9.1 For each of the following files/directories describe the purpose they fulfill 9.2 Your company is about to fire an employee. What steps would you perform to remove the employees account? 9.3 Set up sudo so that a user with the account secretary can run the Linux user management commands which were introduced in this chapter. describe the format of the file The files are /etc/passwd /etc/group /etc/skel

David Jones (20.01.00)

Page 269

85321, Systems Administration

Chapter 10: Managing File Systems

Chapter
Managing File Systems

Introduction
What?
In a previous chapter, we examined the overall structure of the Linux file system. This was a fairly abstract view that didnt explain how the data was physically transferred on and off the disk. Nor in fact, did it really examine the concept of "disks" or even "what" the file system "physically" existed on. In this chapter, we shall look at how Linux interacts with physical devices (not just disks), how in particular Linux uses "devices" with respect to its file system and revisit the Linux file system - just at a lower level.

Why?
Why are you doing this? Doesnt this sound all a bit too like Operating Systems? Unless you are content to accept that all low level interaction with the operating system occurs by a mystical form of osmosis and that you will never have to deal with: A Disk crash - an unfortunate physical event involving one of the read/write heads of a hard disk coming into contact with the platter (which is spinning at high speed) causing the removal of the metallic oxide (the substance that maintains magnetic polarity, thus storing data). This is usually a fatal event for the disk (and sometimes its owner). Adding a disk, mouse, modem terminal or a sound card - unlike some unmentionable operating systems, Linux is not "plug-and-pray". The addition of such a device requires modifications to the system. The accidental erasure of certain essential things called "device files" while the accidental erasure of any file is a traumatic event, the erasure of a device file calls for special action. Installing or upgrading to a kernel or OS release - you may suddenly find that your system doesnt know how to talk to certain things (like your CDROM, your console or maybe your SCSI disk...) - you will need to find out how to solve these problems. Running out of some weird thing called "I-Nodes" - an event which means you cant create any more files.

... then you will definitely need to read this chapter!

David Jones (20.01.00)

Page 270

85321, Systems Administration

Chapter 10: Managing File Systems

Other Resources
Other material discussing file system related information includes HOWTOs CD Writing HOWTO, CDROM HOWTO, Diskless HOWTO, Jaz Drive HOWTO, Large Disk HOWTO, Multi-Disk HOWTO, Optical Disk HOWTO, Root RAID HOWTO, Software RAID HOWTO, UMSDOS HOWTO, Ext2fs Undeletion mini-HOWTO, Hard Disk Upgrade miniHOWTO, Large Disk mini-HOWTO, Partition mini-HOWTO, Partition Rescue mini-HOWTO, Quota mini-HOWTO Guides The Linux Installation and Getting Started Guide includes a section on partitioning and preparing a disk for the installation of Linux. The Linux Systems Administrators Guides chapter 4 provides good coverage of using disks and other storage media. Linux Gazette A free magazine distributed as part of the LDP, issue 21 of the Linux Gazette includes the article A non-technical look inside the Ext2 file system Web resources http://step.polymtl.ca/~ldd/ext2fs/ext2fs_toc.html provides a more indepth technical look at the ext2 file system. The ext2 home page is located at
http://web.mit.edu/tytso/www/linux/ext2.html

A scenario
As we progress through this chapter, we will apply the information to help us solve problems associated with a very common System Administrator's task installing a new hard disk. Our scenario is this: Our current system has a single hard disk and it only has 10% space free (on a good day). This is causing various problems (which we will discuss during the course of this chapter) - needless to say that it is the user directories (off /home) that are using the most space on the system. As our IT department is very poor (we work in a university), we have been budgeting for a new hard disk for the past two years - we had bought a new one a year ago but someone drove a forklift over it. The time has finally arrived - we have a brand new 2.5 gigabyte disk (to complement our existing 500 megabyte one). The size of the disk talked about here should give you some idea of how old this particular chapter is (about 3 years old). Even though todays disks are much larger the basic ideas still apply. How do we install it? What issues should we consider when determining its use?

David Jones (20.01.00)

Page 271

85321, Systems Administration

Chapter 10: Managing File Systems

Devices - Gateways to the kernel

A device is...
A device is just a generic name for any type of physical or logical system component that the operating system has to interact with (or "talk" to). Physical devices include such things as hard disks, serial devices (such as modems, mouse(s) etc.), CD-ROMs, sound cards and tape-backup drives. Logical devices include such things as virtual terminals [every user is allocated a terminal when they log in - this is the point at which output to the screen is sent (STDOUT) and keyboard input is taken (STDIN)], memory, the kernel itself and network ports.

Device files are...

Device files are special types of "files" that allow programs to interact with devices via the OS kernel. These "files" (they are not actually real files in the sense that they do not contain data) act as gateways or entry points into the kernel or kernel related "device drivers".

Device drivers are...

Device drivers are coded routines used for interacting with devices. They essentially act as the "go between" for the low level hardware and the kernel/user interface. Device drivers may be physically compiled into the kernel (most are) or may be dynamically loaded in memory as required.

/dev
/dev is the location where most device files are kept. A listing of /dev will output the names of hundreds of files. The following is an edited extract from the MAKEDEV (a Linux program for making device files - we will examine it later) man page on some of the types of device file that exist in /dev: Take a look at the man page for MAKEDEV on your system for an updated view of this information. Most of it will still be the same. std Standard devices. These include mem - access to physical memory; kmem - access to kernel virtual memory;null - null device; port - access to I/O ports; Virtual Terminals This are the devices associated with the console. This is the virtual terminal tty_, where can be from 0 though 63. Serial Devices Serial ports and corresponding dialout device. For device ttyS_, there is also the device cua_ which is used to dial out with.

David Jones (20.01.00)

Page 272

85321, Systems Administration

Chapter 10: Managing File Systems

Pseudo Terminals (Non-Physical terminals) The master pseudo-terminals are pty[p-s][0-9af] and the slaves are tty[p-s][0-9a-f]. Parallel Ports Standard parallel ports. The devices are lp0, lp1, and lp2. These correspond to ports at 0x3bc, 0x378 and 0x278. Hence, on some machines, the first printer port may actually be lp1. Bus Mice The various bus mice devices. These include: logimouse (Logitech bus mouse), psmouse (PS/2-style mouse), msmouse (Microsoft Inport bus mouse) and atimouse (ATI XL bus mouse) and jmouse (J-mouse). Joystick Devices Joystick. Devices js0 and js1. Disk Devices Floppy disk devices. The device fd_ is the device which autodetects the format, and the additional devices are fixed format (whose size is indicated in the name). The other devices are named as fd___. The single letter _ identifies the type of floppy disk (d = 5.25" DD, h = 5.25" HD, D = 3.5" DD, H = 3.5" HD, E = 3.5" ED). The number _ represents the capacity of that format in K. Thus the standard formats are fd_d360_ fd_h1200_ fd_D720_ fd_H1440_ and fd_E2880_ Devices fd0_ through fd3_ are floppy disks on the first controller, and devices fd4_ through fd7_ are floppy disks on the second controller. Hard disks. The device hdx provides access to the whole disk, with the partitions being hdx[0-20]. The four primary partitions are hdx1 through hdx4, with the logical partitions being numbered from hdx5 though hdx20. (A primary partition can be made into an extended partition, which can hold 4 logical partitions). Drives hda and hdb are the two on the first controller. If using the new IDE driver (rather than the old HD driver), then hdc and hdd are the two drives on the secondary controller. These devices can also be used to access IDE CD-ROMs if using the new IDE driver. SCSI hard disks. The partitions are similar to the IDE disks, but there is a limit of 11 logical partitions (sd_5 through sd_15). This is to allow there to be 8 SCSI disks. Loopback disk devices. These allow you to use a regular file as a block device. This means that images of file systems can be mounted, and used as normal. There are 8 devices, loop0 through loop7.

Tape Devices SCSI tapes. These are the rewinding tape devicest_ and the non-rewinding tape device nst_. QIC-80 tapes. The devices are rmt8, rmt16, tape-d, and tape-reset.

David Jones (20.01.00)

Page 273

85321, Systems Administration

Chapter 10: Managing File Systems

Floppy driver tapes (QIC-117). There are 4 methods of access depending on the floppy tape drive. For each of access methods 0, 1, 2 and 3, the devices rft_ (rewinding) and nrft_ (non-rewinding) are created. CD-ROM Devices SCSI CD players. Sony CDU-31A CD player. Mitsumi CD player. Sony CDU-535 CD player. LMS/Philips CD player. Sound Blaster CD player. The kernel is capable of supporting 16 CDROMs, each of which is accessed as sbpcd[0-9a-f]. These are assigned in groups of 4 to each controller. Audio These are the audio devices used by the sound driver. These include mixer, sequencer, dsp, and audio. Devices for the PC Speaker sound driver. These are pcmixer. pxsp, and pcaudio. Miscellaneous Generic SCSI devices. The devices created are sg0 through sg7. These allow arbitrary commands to be sent to any SCSI device. This allows for querying information about the device, or controlling SCSI devices that are not one of disk, tape or CD-ROM (e.g. scanner, writable CD-ROM).

While the /dev directory contains the device files for many types of devices, only those devices that have device drivers present in the kernel can be used. For example, while your system may have a /dev/sbpcd, it doesnt mean that your kernel can support a Sound Blaster CD. To enable the support, the kernel will have to be recompiled with the Sound Blaster driver included - a process we will examine in a later chapter.

Physical characteristics of device files

If you were to examine the output of the ls -al command on a device file, youd see something like:
psyche:~/sanotes$ ls -al /dev/console crw--w--w1 jamiesob users 4, 0 Mar 31 09:28 /dev/console

In this case, we are examining the device file for the console. There are two major differences in the file listing of a device file from that of a "normal" file, for example:
psyche:~/sanotes$ ls -al iodev.html -rw-r--r-1 jamiesob users 7938 Mar 31 12:49 iodev.html

The first difference is the first character of the "file permissions" grouping this is actually the file type. On directories this is a "d", on "normal" files it will be blank but on devices it will be "c" or "b". This character indicates c for

David Jones (20.01.00)

Page 274

85321, Systems Administration

Chapter 10: Managing File Systems

character mode or b for block mode. This is the way in which the device interacts - either character by character or in blocks of characters. For example, devices like the console output (and input) character by character. However, devices like hard disks read and write in blocks. You can see an example of a block device by the following:
psyche:~/sanotes$ ls -al /dev/hda brw-rw---1 root disk 3, 0 Apr 28 1995 /dev/hda

(hda is the first hard drive) The second difference is the two numbers where the file size field usually is on a normal file. These two numbers (delimited by a comma) are the major and minor device numbers.

Major and minor device numbers are...

Major and minor device numbers are the way in which the kernel determines which device is being used, therefore what device driver is required. The kernel maintains a list of its available device drivers, given by the major number of a device file. When a device file is used (we will discuss this in the next section), the kernel runs the appropriate device driver, passing it the minor device number. The device driver determines which physical device is being used by the minor device number. For example:
psyche:~/sanotes$ ls -al /dev/hda brw-rw---1 root disk psyche:~/sanotes$ ls -al /dev/hdb brw-rw---1 root disk 3, 3, 0 Apr 28 64 Apr 28 1995 /dev/hda 1995 /dev/hdb

What this listing shows is that a device driver, major number 3, controls both hard drives hda and hdb. When those devices are used, the device driver will know which is which (physically) because hda has a minor device number of 0 and hdb has a minor device number of 64.

Finding the devices on your system

The /proc file system provides access to certain values within the kernel of the operating system. This means you cant copy files into the /proc directory, most of the directory structure is read only. Instead you read the files under /proc to find out information about the configuration of your system and in particular the kernel. For example, the file /proc/cpuinfo contains information about the CPU of your system.
[root@faile processor vendor_id cpu family model model name stepping cpu MHz fdiv_bug hlt_bug sep_bug f00f_bug /root]# cat /proc/cpuinfo : 0 : GenuineIntel : 5 : 8 : Mobile Pentium MMX : 1 : 233.867806 : no : no : no : yes

David Jones (20.01.00)

Page 275

85321, Systems Administration

Chapter 10: Managing File Systems

coma_bug : fpu : fpu_exception cpuid level : wp : flags : bogomips :

no yes : yes 1 yes fpu vme de pse tsc msr mce cx8 mmx 466.94

One of the other files in the proc file system is called devices. As you might expect it contains a list of the devices for which device drivers exist in your machines kernel.
[root@faile /root]# cat /proc/devices Character devices: 1 mem 2 pty 3 ttyp 4 ttyS 5 cua 7 vcs 10 misc 14 sound 29 fb 36 netlink 128 ptm 136 pts 162 raw 254 pcmcia Block devices: 1 ramdisk 2 fd 3 ide0 9 md 22 ide1

Many UNIX commands use the /proc file system including ps, top and uptime. The procinfo command is another useful command for displaying system status information from /proc.
[root@faile /root]# procinfo Linux 2.2.12-20 ([email protected]) (gcc egcs-2.91.66) #1 1CPU [faile] Memory: Total Used Free Shared Buffers Cached Mem: 127948 124180 3768 108864 7336 61144 Swap: 72252 128 72124 Bootup: Wed Jan 12 08:52:02 2000 user : 4018w nice : system: idle : uptime: irq [0] irq irq irq irq 0: 1: 2: 3: 4: 0:03:05.92 0:00:00.07 0:00:47.27 0:52:11.09 0:56:04.34 336435 timer 12419 keyboard 0 cascade [4] 7532 3c589_cs 5 5.5% 0.0% 1.4% 93.1% Load average: 0.02 0.06 0.07 1/88 1034 page in : page out: swap in : swap out: context : irq 24920 6665 1 32 621650 7: 0 MSS audio codec 1 4 14187 1 rtc i82365 PS/2 Mouse fpu disk 1: 5274r

irq 8: irq 11: irq 12: irq 13:

David Jones (20.01.00)

Page 276

85321, Systems Administration

Chapter 10: Managing File Systems

irq irq

5: 6:

2 3

irq 14: irq 15:

239393 ide0 33230 ide1

Device files and Device drivers The presence of a device file on your system does not mean you can actually use that device. You also need the device driver. The content of the file /proc/devices is the list of device drivers in your kernel. To use a particular device you need to have both the device driver and the device file. Remember, the device file is simply an interface to the device driver. The device file doesnt know how to talk to the device. For example, my laptop computer has all the device files for SCSI hard drives (/dev/sda1 /dev/sda2 etc). However, I still cant use SCSI hard-drives with the laptop because the kernel for Linux does not contain any device drivers for SCSI drives. Look at the content of the /proc/devices file in the previous example.

Why use device files?

It may seem using files is a roundabout method of accessing devices - what are the alternatives? Other operating systems provide system calls to interact with each device. This means that each program needs to know the exact system call to talk to a particular device. With UNIX and device files, this need is removed. With the standard open, read, write, append etc. system calls (provided by the kernel), a program may access any device (transparently) while the kernel determines what type of device it is and which device driver to use to process the call. [You will remember from Operating Systems that system calls are the services provided by the kernel for programs.] Using files also allows the system administrator to set permissions on particular devices and enforce security - we will discuss this in detail later. The most obvious advantage of using device files is shown by the way in which as a user, you can interact with them. For example, instead of writing a special program to play .AU sound files, you can simply:
psyche:~/sanotes$ cat test.au > /dev/audio

This command pipes the contents of the test.au file into the audio device. Two things to note: 1) This will only work for systems with audio (sound card) support compiled into the kernel (i.e. device drivers exist for the device file) and 2) this will only work for .AU files - try it with a .WAV and see (actually, listen) what happens. The reason for this is that .WAV (a Windows audio format) has to be interpreted first before it can be sent to the sound card. You will not probably need to be the root user to perform the above command as the /dev/audio device has write permissions to all users. However, dont cat anything to a device unless you know what you are doing - we will discuss why later.

David Jones (20.01.00)

Page 277

85321, Systems Administration

Chapter 10: Managing File Systems

Creating device files

There are two ways to create device files - the easy way or the hard way! The easy way involves using the Linux command MAKEDEV. This is actually a script that can be found in the /dev directory. MAKEDEV accepts a number of parameters (you can check what they are in the man pages. In general, MAKEDEV is run as:
/dev/MAKEDEV device

where device is the name of a device file. If for example, you accidentally erased or corrupted your console device file (/dev/console) then youd recreate it by issuing the commend:
/dev/MAKEDEV console

NOTE! This must be done as the root user However, what if your /dev directory had been corrupted and you lost the MAKEDEV script? In this case youd have to manually use the mknod command. With the mknod command you must know the major and minor device number as well as the type of device (character or block). To create a device file using mknod, you issue the command:
mknod device_file_name device_type major_number minor_number

For example, to create the device file for COM1 a.k.a. /dev/ttys0 (usually where the mouse is connected) youd issue the command:
mknod /dev/ttyS0 c 4 240

Ok, so how do you know what type a device file is and what major and minor number it has so you can re-create it? The scouting (or is that the cubs?) solution to every problem in the world, be prepared, comes into play. Being a good system administrator, youd have a listing of every device file stored in a file kept safely on disk. Youd issue the command:
ls -al /dev > /mnt/device_file_listing

before you lost your /dev directory in a cataclysmic disaster, so you could read the file and recreate the /dev structure (it might also be smart to copy the MAKEDEV script onto this same disk just to make your life easier :). is only found on Linux systems. It relies on the fact that the major and minor devices numbers for the system are hard-coded into the script - running MAKEDEV on a non-Linux system wont work because:
MAKEDEV

The device names are different The major and minor numbers of similar devices are different Note however that similar scripts to MAKEDEV can be found on most modern versions of UNIX.

David Jones (20.01.00)

Page 278

85321, Systems Administration

Chapter 10: Managing File Systems

The use and abuse of device files

Device files are used directly or indirectly in every application on a Linux system. When a user first logs in, they are assigned a particular device file for their terminal interaction. This file can be determined by issuing the command:
tty

For example:
psyche:~/sanotes$ tty /dev/ttyp1 psyche:~/sanotes$ ls -al /dev/ttyp1 crw------1 jamiesob tty4, 193 Apr 2 21:14 /dev/ttyp1

Notice that as a user, I actually own the device file! This is so I can write to the device file and read from it. When I log out, it will be returned to:
c--------/dev/ttyp1 1 root root 4, 193 Apr 2 20:33

Try the following:

read X < /dev/ttyp1 ; echo "I wrote $X" echo "hello there" > /dev/ttyp1

You should see something like:

psyche:~/sanotes$ read X < /dev/ttyp1 ; echo "I wrote $X" hello I wrote hello psyche:~/sanotes$ echo "hello there" > /dev/ttyp1 hello there

A very important device file is that which is assigned to your hard disk. In my case /dev/hda is my primary hard disk, its device file looks like:
brw-rw---1 root disk 3, 0 Apr 28 1995 /dev/hda

Note that as a normal user, I cant directly read and write to the hard disk device file - why do you think this is? Reading and writing to the hard disk is handled by an intermediary called the file system. We will examine the role of the file system in later sections, but for the time being, you should be aware that the file system decides how to use the disk, how to find data and where to store information about what is on the disk. Bypassing the file system and writing directly to the device file is a very dangerous thing - device drivers have no concept of file systems, files or even the data that is stored in them; device drivers are only interested in reading and writing chunks of data (called blocks) to physical sectors of the disk. For example, by directly writing a data file to a device file, you are effectively instructing the device driver to start writing blocks of data onto the disk from where ever the disk head was sitting! This can (depending on which sector and track the disk was set to) potentially wipe out the entire file structure, boot sector and all the data. Not a good idea to try it. NEVER should you issue a command like:

David Jones (20.01.00)

Page 279

85321, Systems Administration

Chapter 10: Managing File Systems

cat some_file > /dev/hda1

As a normal user, you cant do this - but you can as root! Reading directly from the device file is also a problem. While not physically damaging the data on the disk, by allowing users to directly read blocks, it is possible to obtain information about the system that would normally be restricted to them. For example, was someone clever enough to obtain a copy of the blocks on the disk where the shadow password file resided (a file normally protected by file permissions so users can view it), they could potentially reconstruct the file and run it through a crack program. Exercises
10.1.

Use the tty command to find out what device file you are currently logged in from. In your home directory, create a device file called myterm that has the same major and minor device number. Log into another session and try redirecting output from a command to myterm. What happens? Use the tty command to find out what device file you are currently logged in on. Try using redirection commands to read and write directly to the device. With another user (or yourself in another session) change the permissions on the device file so that the other user can write to it (and you to theirs). Try reading and writing from each others device files. Log into two terminals as root. Determine the device file used by one of the sessions, take note of its major and minor device number. Delete the device file - what happens to that session. Log out of the session - now what happens? Recreate the device file.

10.2.

10.3.

Devices, Partitions and File systems

Device files and partitions
Apart from general device files for entire disks, individual device files for partitions exist. These are important when trying to understand how individual "parts" of a file hierarchy may be spread over several types of file system, partitions and physical devices. Partitions are non-physical (I am deliberately avoiding the use of the word "logical" because this is a type of partition) divisions of a hard disk. IDE Hard disks may have 4 primary partitions, one of which must be a boot partition if the hard disk is the primary (modern systems have primary and secondary disk controllers) master (first hard disk) [this is the partition BIOS attempts to load a bootstrap program from at boot time]. Each primary partition can be marked as an extended partition which can be further divided into four logical partitions. By default, Linux provides device files for the four primary partitions and 4 logical partitions per primary/extended partition. For example, a listing of the device files for my primary master hard disk reveals:

David Jones (20.01.00)

Page 280

85321, Systems Administration

Chapter 10: Managing File Systems

brw-rw---brw-rw---brw-rw---brw-rw---brw-rw---brw-rw---brw-rw---brw-rw---brw-rw---brw-rw---brw-rw---brw-rw---brw-rw---brw-rw---brw-rw---brw-rw---brw-rw----

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

root root root root root root root root root root root root root root root root root

disk disk disk disk disk disk disk disk disk disk disk disk disk disk disk disk disk

3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,

0 1 10 11 12 13 14 15 16 2 3 4 5 6 7 8 9

Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr Apr

28 28 28 28 28 28 28 28 28 28 28 28 28 28 28 28 28

1995 1995 1995 1995 1995 1995 1995 1995 1995 1995 1995 1995 1995 1995 1995 1995 1995

/dev/hda /dev/hda1 /dev/hda10 /dev/hda11 /dev/hda12 /dev/hda13 /dev/hda14 /dev/hda15 /dev/hda16 /dev/hda2 /dev/hda3 /dev/hda4 /dev/hda5 /dev/hda6 /dev/hda7 /dev/hda8 /dev/hda9

Partitions are usually created by using a system utility such as fdisk. Generally fdisk will ONLY be used when a new operating system is installed or a new hard disk is attached to a system. Our existing hard disk would be /dev/hda1 (we will assume that we are using an IDE drive, otherwise wed be using SCSI devices /dev/sd*). Our new hard disk (well make it a slave to the first) will be
/dev/hdb1.

Partitions and file systems

Every partition on a hard disk has an associated file system (the file system type is actually set when fdisk is run and a partition is created). For example, in DOS machines, it was usual to devote the entire hard disk (therefore the entire disk contained one primary partition) to the FAT (File Allocation Table) based file system. This is generally the case for most modern operating systems including Windows 95, Win NT and OS/2. However, there are occasions when you may wish to run multiple operating systems off the one disk; this is when a single disk will contain multiple partitions, each possibly containing a different file system. With UNIX systems, it is normal procedure to use multiple partitions in the file system structure. It is quite possible that the file system structure is spread over multiple partitions and devices, each a different "type" of file system. What do I mean by "type" of file system? Linux can support (or "understand", access, read and write to) many types of file systems including: minix, ext, ext2, umsdos, msdos, proc, nfs, iso9660, xenix, Sysv, coherent, hpfs. (There is also support for the Windows 95 and Win NT file system). A file system is simply a set or rules and algorithms for accessing files. Each system is different; one file system cant read the other. Like device drivers, file systems are compiled into the kernel - only file systems compiled into the kernel can be accessed by the kernel.

David Jones (20.01.00)

Page 281

85321, Systems Administration

Chapter 10: Managing File Systems

To discover what file systems your system supports, you can display the contents of the /proc/filesystems file. On our new disk, if we were going to use a file system that was not supported by the kernel, we would have to recompile the kernel at this point.

Partitions and Blocks

The smallest unit of information that can be read from or written to a disk is a block. Blocks cant be split up - two files cant use the same block, therefore even if a file only uses one byte of a block, it is still allocated the entire block. When partitions are created, the first block of every partition is reserved as the boot block. However, only one partition may act as a boot partition. BIOS checks the partition table of the first hard disk at boot time to determine which partition is the boot partition. In the boot block of the boot partition there exists a small program called a bootstrap loader - this program is executed at boot time by BIOS and is used to launch the OS. Systems that contain two or more operating systems use the boot block to house small programs that ask the user to chose which OS they wish to boot. One of these programs is called lilo and is provided with Linux systems. The second block on the partition is called the superblock. It contains all the information about the partition including information on: The size of the partition The physical address of the first data block The number and list of free blocks Information of what type of file system uses the partition When the partition was last modified

The remaining blocks are data blocks. Exactly how they are used and what they contain are up to the file system using the partition.

Using the partitions

So how does Linux use these partitions and file systems? Linux logically attaches (this process is called mounting) different partitions and devices to parts of the directory structure. For example, a system may have:
/ mounted to /dev/hda1 /usr mounted to /dev/hda2 /home mounted to /dev/hda3 /usr/local mounted to /dev/hda4 /var/spool mounted to /dev/hdb1 /cdrom mounted to /dev/cdrom /mnt mounted to /dev/fd0

Yet to a user of the system, the physical location of the different parts of the directory structure is transparent! How does this work?
David Jones (20.01.00) Page 282

85321, Systems Administration

Chapter 10: Managing File Systems

The Virtual File System

The Linux kernel contains a layer called the VFS (or Virtual File System). The VFS processes all file-oriented IO system calls. Based on the device that the operation is being performed on, the VFS decides which file system to use to further process the call. The exact list of processes that the kernel goes through when a system call is received follows along the lines of: A process makes a system call. The VFS decides what file system is associated with the device file that the system call was made on. The file system uses a series of calls (called Buffer Cache Functions) to interact with the device drivers for the particular device. The device drivers interact with the device controllers (hardware) and the actual required processes are performed on the device.

Figure 15.1 represents this.

The

Figure 15.1 Virtual File System

David Jones (20.01.00)

Page 283

85321, Systems Administration

Chapter 10: Managing File Systems

Dividing up the file hierarchy - why?

Why would you bother partitioning a disk and using different partitions for different directories? The reasons are numerous and include: Separation Issues Different directory branches should be kept on different physical partitions for reasons including: Certain directories will contain data that will only need to be read, others will need to be both read and written. It is possible (and good practice) to mount these partitions restricting such operations. Directories including /tmp and /var/spool can fill up with files very quickly, especially if a process becomes unstable or the system is purposely flooded with email. This can cause problems. For example, let us assume that the /tmp directory is on the same partition as the /home directory. If the /tmp directory causes the partition to be filled no user will be able to write to their /home directory, there is no space. If /tmp and /home are on separate partitions the filling of the /tmp partition will not influence the /home directories. The logical division of system software, local software and home directories all lend themselves to separate partitions

Backup Issues These include: Separating directories like /usr/local onto separate partitions makes the process of an OS upgrade easier - the new OS version can be installed over all partition except the partition that the /usr/local system exists on. Once installation is complete the /usr/local partition can be reattached. The actual size of the partition can make it easier to perform backups - it isnt as easy to backup a single 2.1 Gig partition as it is to backup four 500 Meg partitions. This does depend on the backup medium you are using. Some medium will handle a 2.1 Gb partition quite easily.

Performance Issues By spreading the file system over several partitions and devices, the IO load is spread around. It is then possible to have multiple seek operations occurring simultaneously - this will improve the speed of the system. While splitting the directory hierarchy over multiple partitions does address the above issues, it isnt always that simple. A classic example of this is a system that contained its Web programs and data in the /var/spool directory. Obviously the correct location for this type of program is the /usr branch probably somewhere off the /usr/local system. The reason for this strange
David Jones (20.01.00) Page 284

85321, Systems Administration

Chapter 10: Managing File Systems

location? ALL the other partitions on the system were full or nearly full - this was the only place left to install the software! And the moral of the story is? When partitions are created for different branches of the file hierarchy, the future needs of the system must be considered - and even then, you wont always be able to adhere to what is "the technically correct" location to place software.

Scenario Update
At this point, we should consider how we are going to partition our new hard disk. As given by the scenario, our /home directory is using up a lot of space (we would find this out by using the du command). We have the option of devoting the entire hard disk to the /home structure but as it is a 2.5 Gig disk we could probably afford to divide it into a couple of partitions. As the /var/spool directory exists on the same partition as root, we have a potential problem of our root partition filling up - it might be an idea to separate this. As to the size of the partitions? As our system has just been connected to the Internet, our users have embraced FTP - our /home structure is consuming 200 Megabytes but we expect this to increase by a factor of 10 over the next 2 years. Our server is also receiving increased volumes of email, so our spool directory will have to be large. A split of 2 Gigabytes to 500 Megabytes will probably be reasonable. To create our partitions, we will use the fdisk program. We will create two primary partitions, one of 2 Gigabytes and one of 500 Megabytes - these we will mark as Linux partitions.

The Linux Native File System - ext2

Overview
Historically, Linux has had several native file systems. Originally there was Minix which supported file systems of up to 64 megabytes in size and 14 character file names. With the advent of the virtual file system (VFS) and support for multiple file systems, Linux has seen the development of Ext FS (Extended File System), Xia FS and the current ext2 FS. ext2 (the second extended file system) has longer file names (255 characters), larger file sizes (2 GB) and bigger file system support (4 TB) than any of the existing Linux file systems. In this section, we will examine how ext2 works.

I-Nodes
use a complex but extremely efficient method of organising block allocation to files. This system relies on data structures called I-Nodes. Every file on the system is allocated an I-Node - there can never be more files than I-Nodes.
ext2

David Jones (20.01.00)

Page 285

85321, Systems Administration

Chapter 10: Managing File Systems

This is something to consider when you format a partition and create the file system - you will be asked how many I-Nodes you wish create. Generally, ten percent of the file system should be I-Nodes. This figure should be increased if the partition will contain lots of small files or decreased if the partition will contain few but large files. Figure 15.2 is a graphical representation on an I-Node.

Figure 15.2 I-Node Structure

Typically an I-Node will contain: The owner (UID) and group owner (GID) of the file. The type of file - is the file a directory or another type of special file? User access permissions - which users can do what with the file The number of hard links to the file - the same physical file may be accessed under several names; we will examine how later. The size of the file The time the file was last modified

David Jones (20.01.00)

Page 286

85321, Systems Administration

Chapter 10: Managing File Systems

The time the I-Node was last changed - if permissions or information on the file change then the I-Node is changed. The addresses of 13 data blocks - data blocks are where the contents of the file are placed. A single indirect pointer - this points to a special type of block called a single indirect block. This is a block that contains the addresses of at least 256 other data blocks; the exact number depends of the file system and implementation. A double indirect pointer - this points to a special type of block called a double indirect block. This block points to a number of single indirect blocks. A triple indirect pointer - this points to a special type of block called a triple indirect block. This block points to a number of double indirect blocks.

Using this system, ext2 can cater for a file two gigabytes in size! However, just because an I-Node can access all those data blocks doesnt mean that they are automatically allocated to the file when it is created - obviously! As the file grows, blocks are allocated, starting with the first direct 13 data blocks, then moving on to the single indirect blocks, then to the double, then to the triple. Note that the actual name of the file is not stored in the I-Node. This is because the names of files are stored in directories, which are themselves files.

Physical Structure and Features

uses a decentralised file system management scheme involving a "block group" concept. What this means is that the file systems are divided into a series of logical blocks. Each block contains a copy of critical information about the file systems (the super block and information about the file system) as well as an I-Node, and data block allocation tables and blocks. Generally, the information about a file (the I-Node) will be stored close to the data blocks. The entire system is very robust and makes file system recovery less difficult. The ext2 file system also has some special features which make it stand out from existing file systems including: Logical block size - the size of data blocks can be defined when the file system is created; this is not dependent on physical data block size.
ext2

File system state checks - the file system keeps track of how many times it was "mounted " (or used) and what state it was left in at the last shutdown. The file system reserves 5% of the file system for the root user - this means that if a user program fills a partition, the partition is still useable by root (for recovery) because there is reserve space.

A more comprehensive description of the ext2 file system can be found at http://web.mit.edu/tytso/www/linux/ext2.html .

David Jones (20.01.00)

Page 287

85321, Systems Administration

Chapter 10: Managing File Systems

dumpe2fs Command The dumpe2fs command is useful for providing information about the super block and blocks group of a file system.
[root@faile /root]# dumpe2fs /dev/hda1 dumpe2fs 1.15, 18-Jul-1999 for EXT2 FS 0.5b, 95/08/09 Filesystem volume name: <none> Last mounted on: <not available> Filesystem UUID: 459a17c0-7b5c-11d3-93ed-b8c49e459f04 Filesystem magic number: 0xEF53 Filesystem revision #: 1 (dynamic) Filesystem features: sparse_super Filesystem state: not clean Errors behavior: Continue Filesystem OS type: Linux Inode count: 6024 Block count: 24066 Reserved block count: 1203 Free blocks: 20655 Free inodes: 5999 First block: 1 Block size: 1024 Fragment size: 1024 Blocks per group: 8192 Fragments per group: 8192 Inodes per group: 2008 Inode blocks per group: 251 Last mount time: Wed Jan 12 08:52:22 2000 Last write time: Wed Jan 12 08:52:26 2000 Mount count: 20 Maximum mount count: 20 Last checked: Sat Jan 1 08:35:59 2000 Check interval: 15552000 (6 months) Next check after: Thu Jun 29 08:35:59 2000 Reserved blocks uid: 0 (user root) Reserved blocks gid: 0 (group root) First inode: 11 Inode size: 128

Group 0: (Blocks 1 -- 8192) Block bitmap at 3 (+2), Inode bitmap at 4 (+3) Inode table at 5 (+4) 5290 free blocks, 1983 free inodes, 2 directories Free blocks: 2901-2902, 2905-8192 Free inodes: 26-2008 Group 1: (Blocks 8193 -- 16384) Block bitmap at 8195 (+2), Inode bitmap at 8196 (+3) Inode table at 8197 (+4) 7937 free blocks, 2008 free inodes, 0 directories Free blocks: 8448-16384 Free inodes: 2009-4016 Group 2: (Blocks 16385 -- 24065) Block bitmap at 16385 (+0), Inode bitmap at 16386 (+1) Inode table at 16389 (+4) 7428 free blocks, 2008 free inodes, 0 directories Free blocks: 16387-16388, 16640-24065 Free inodes: 4017-6024

David Jones (20.01.00)

Page 288

85321, Systems Administration

Chapter 10: Managing File Systems

Creating file systems

mkfs
Before a partition can be mounted (or used), it must first have a file system installed on it - with ext2, this is the process of creating I-Nodes and data blocks. This process is the equivalent of formatting the partition (similar to MSDOSs "format" command). Under Linux, the command to create a file system is called mkfs. The command is issued in the following way:
mkfs [-c] [ -t fstype ] filesys [ blocks ] eg. mkfs -t ext2 /dev/fd0 # Make a ext2 file system on a disk

where: -c forces a check for bad blocks -t fstype specifies the file system type filesys is either the device file associated with the partition or device OR is the directory where the file system is mounted (this is used to erase the old file system and create a new one) blocks specifies the number of blocks on the partition to allocate to the file system Be aware that creating a file system on a device with an existing file system will cause all data on the old file system to be erased.

Scenario Update
Having partitioned our disk, we must now install a file system on each partition. is the logical choice. Be aware that this wont always be the case and you should educate yourself on the various file systems available before making a choice.
ext2

Assuming /dev/hdb1 is the 2GB partition and /dev/hdb2 is the 500 MB partition, we can create ext2 file systems using the commands:
mkfs -t ext2 -c /dev/hdb1 mkfs -t ext2 -c /dev/hdb2

This assumes the default block size and the default number of I-Nodes. If we wanted to be more specific about the number of I-Nodes and block size, we could specify them. mkfs actually calls other programs to create the file system - in the ext2 case, mke2fs. Generally, the defaults are fine - however, if we knew that we were only storing a few large files on a partition, then wed reduce the I-Node to data block ratio. If we knew that we were storing lots of small files on a partition, wed increase the I-Node to data block ration and

David Jones (20.01.00)

Page 289

85321, Systems Administration

Chapter 10: Managing File Systems

probably decrease the size of the data blocks (there is no point using 4K data blocks when the file size average is around 1K). Exercises
10.4.

Create an ext2 file system on a floppy disk using the defaults. Create a ext2 file system on a floppy disk with only one I-node. How much disk space do you save using only 1 I-node? What restriction does this place on the floppy disk?

Mounting and UN-mounting Partitions and Devices

Mount
To attach a partition or device to part of the directory hierarchy you must mount its associated device file. To do this, you must first have a mount point - this is simply a directory where the device will be attached. This directory will exist on a previously mounted device (with the exception of the root directory (/) which is a special case) and will be empty. If the directory is not empty, then the files in the directory will no longer be visible while the device to mounted to it, but will reappear after the device has been disconnected (or unmounted). To mount a device , you use the mount command:
mount [switches] device_file mount_point

With some devices, mount will detect what type of file system exists on the device, however it is more usual to use mount in the form of:
mount [switches] -t file_system_type device_file mount_point

Generally, only the root user can use the mount command - mainly due to the fact that the device files are owned by root. For example, to mount the first partition on the second hard drive off the /usr directory and assuming it contained the ext2 file system youd enter the command:
mount -t ext2 /dev/hdb1 /usr

A common device that is mounted is the floppy drive. A floppy disk generally contains the msdos file system (but not always) and is mounted with the command:
mount -t msdos /dev/fd0 /mnt

Note that the floppy disk was mounted under the /mnt directory? This is because the /mnt directory is the usual place to temporally mount devices. To see what devices you currently have mounted, simply type the command mount. Typing it on my system reveals:
/dev/hda3 on / type ext2 (rw) /dev/hda1 on /dos type msdos (rw) none on /proc type proc (rw)

David Jones (20.01.00)

Page 290

85321, Systems Administration

Chapter 10: Managing File Systems

/dev/cdrom on /cdrom type iso9660 (ro) /dev/fd0 on /mnt type msdos (rw)

Each line tells me what device file is mounted, where it is mounted, what file system type each partition is and how it is mounted (ro = read only, rw = read/write). Note the strange entry on line three - the proc file system? This is a special "virtual" file system used by Linux systems to store information about the kernel, processes and current resource usages. It is actually part of the systems memory - in other words, the kernel sets aside an area of memory which it stores information about the system in - this same area is mounted onto the file system so user programs can easily gain this information. To release a device and disconnect it from the file system, the umount command is used. It is issued in the form:
umount device_file or umount mount_point

For example, to release the floppy disk, youd issue the command:
umount /mnt or umount /dev/fd0

Again, you must be the root user or a user with privileges to do this. You cant unmount a device/mount point that is in use by a user (the users current working directory is within the mount point) or is in use by a process. Nor can you unmount devices/mount points which in turn have devices mounted to them. All of this begs the question - how does the system know which devices to mount when the OS boots?

Mounting with the /etc/fstab file

In true UNIX fashion, there is a file which governs the behaviour of mounting devices at boot time. In Linux, this file is /etc/fstab. But there is a problem - if the fstab file lives in the /etc directory (a directory that will always be on the root partition (/)), how does the kernel get to the file without first mounting the root partition (to mount the root partition, you need to read the information in the /etc/fstab file!)? The answer to this involves understanding the kernel (a later chapter) - but in short, the system cheats! The kernel is "told" (how it is told doesnt concern us yet) on which partition to find the root file system; the kernel mounts this in read only mode, assuming the Linux native ext2 file system, then reads the fstab file and re-mounts the root partition (and others) according to instructions in the file. So what is in the file? An example line from the fstab file uses the following format:
device_file mount_point file_system_type mount_options [n] [n]

The first three fields are self explanatory; the fourth field, mount_options defines how the device will be mounted (this includes information of access mode ro/rw, execute permissions and other information) - information on this can be found in the mount man pages (note that this field usually contains the word "defaults"). The fifth and sixth fields will usually either not be included
David Jones (20.01.00) Page 291

85321, Systems Administration

Chapter 10: Managing File Systems

or be "1" - these two fields are used by the system utilities dump and fsck respectively - see the man pages for details. As an example, the following is my /etc/fstab file:
/dev/hda5 /dev/hda1 /dev/cdrom /dev/hda6 /dev/fd0 none none / /boot /mnt/cdrom swap /mnt/floppy /proc /dev/pts ext2 ext2 iso9660 swap ext2 proc devpts defaults defaults noauto,owner,ro defaults noauto,owner defaults gid=5,mode=620 1 1 0 0 0 0 0 1 2 0 0 0 0 0

This is a fairly standard /etc/fstab file created by installing Redhat 6.1. Linux itself has two main partitions which are used to store files, / and /boot. /boot is a small partition which contains the kernel, LILO configuration and a small number of files required to get the system going. The / file system contains every other file on my system. While this is somewhat okay for a home system it would probably be better to split this partition up based on some of the guidelines discussed in this chapter. The swap partition is required by Linux but doesnt actually contain files. It is used directly by the kernel. You can see entries for the CD-ROM and floppy drive of the system and for mounting them under the standard /mnt directory. Lastly there are the two pseudo file systems proc and pts.

Scenario Update
The time has come for us to use our partitions. The following procedure should be followed: Mount each partition (one at a time) off /mnt Eg.
mount -t ext2 -o defaults /dev/hdb1 /mnt

Copy the files from the directory that is going to reside on the partition TO the partition Eg.
cp - a /home /mnt

Modify the /etc/fstab file to mount the partition off the correct directory Eg.
/dev/hdb1 /home ext2 defaults 1 1

Test your changes by rebooting and using the partition Unmount the partition and remove the old files (or back them up).
umount /home rm -r /home mount -t ext2 -o defaults /dev/hdb1 /home

The new hard disk should be now installed and configured correctly! Exercises
10.5. 10.6.

Mount a floppy disk under the /mnt/floppy directory. Carefully examine your /etc/fstab file - work out what each entry means.

David Jones (20.01.00)

Page 292

85321, Systems Administration

Chapter 10: Managing File Systems

10.7.

Change to the /mnt/floppy directory (while the disk is mounted) - now try to unmount the disk - does this work? Why/Why not?

File Operations
Creating a file
When a file is created, the following process is performed: An I-Node is allocated to the file. (If one is available) An entry is added to the current directory - remember, the directory is a file itself. This entry contains the name of the file and a pointer to I-Node used by the file. The link count on the files I-Node is set to 1 (any I-Node with a link count of 0 is not in use). Any blocks required to store the file contents are allocated.

Linking files
As we have previously encountered, there are occasions when you will want to access a file from several locations or by several names. The process of doing this is called linking. There are two methods of doing this - Hard Linking and Soft Linking. Hard Links are generated by the following process: An entry is added to the current directory with the name of the link together with a pointer to the I-Node used by the original file. The I-Node of the original file is updated and the number of files linked to it is incremented.

Soft Links are generated by the following process: An I-Node is allocated to the soft link file - the type of file is set to softlink. An entry is added to the current directory with the name of the link together with a pointer to the allocated I-Node. A data block is allocated for the link in which is placed the name of the original file.

Programs accessing a soft link cause the file system to examine the location of the original (linked-to) file and then carry out operations on that file. The following should be noted about links: Hard links may only be performed between files on the same physical partition - the reason for this is that I-Nodes pointers can only point to INodes of the same partition Any operation performed on the data in link is performed on the original file. Any chmod operations performed on a hard link are reflected on both the hard link file and the file it is linked to. chmod operations on soft links are
Page 293

David Jones (20.01.00)

85321, Systems Administration

Chapter 10: Managing File Systems

reflected on the original file but not on the soft link - the soft link will always have full file permissions (lrwxrwxrwx) . So how do you perform these mysterious links?

ln
The command for both hard and soft link files is ln. It is executed in the following way:
ln source_file link_file_name # Hard Links or ln -s source_file link_file_name# Soft Links

For example, look at the following operations on links: Create the file and check the ls listing:
psyche:~$ touch base psyche:~$ ls -al base -rw-r--r-1 jamiesob users

0 Apr

5 17:09 base

Create a soft link and check the ls listing of it and the original file
psyche:~$ ln psyche:~$ ls lrwxrwxrwx psyche:~$ ls -rw-r--r--s base softbase -al softbase 1 jamiesob users -al base 1 jamiesob users

4 Apr 0 Apr

5 17:09 softbase -> base 5 17:09 base

Create a hard link and check the ls listing of it, the soft link and the original file
psyche:~$ ln base hardbase psyche:~$ ls -al hardbase -rw-r--r-2 jamiesob users 0 Apr psyche:~$ ls -al base -rw-r--r-2 jamiesob users 0 Apr psyche:~$ ls -il base 132307 -rw-r--r-2 jamiesob users psyche:~$ ls -il softbase 132308 lrwxrwxrwx 1 jamiesob users psyche:~$ ls -il hardbase 132307 -rw-r--r-2 jamiesob users

5 17:09 hardbase 5 17:09 base 0 Apr 4 Apr 0 Apr 5 17:09 base 5 17:09 softbase ->base 5 17:09 hardbase

Note the last three operations (checking the I-Node number) - see how the hard link shares the I-Node of the original file? Links are removed by simply deleting the link with the rm (or on non-Linux systems unlink) command. Note that deleting a file that has soft links is different to deleting a file with hard links - deleting a soft-linked file causes the I-Node (thus data blocks) to be deallocated - no provision is made for the soft link which is now "pointing" to a file that doesnt exist. However, a file with hard links to it has its entry removed from the directory, but neither its I-Node nor data blocks are deallocated - the link count on the INode is simply decremented. The I-Node and data blocks will only be deallocated when there are no other files hard linked to it. Exercises
10.8.

Locate all files on the system that are soft links (Hint: use find).

David Jones (20.01.00)

Page 294

85321, Systems Administration

Chapter 10: Managing File Systems

Checking the file system

Why Me?
It is a sad truism that anything that can go wrong will go wrong - especially if you dont have backups! In any event, file system "crashes" or problems are an inevitable fact of life for a System Administrator. Crashes of a non-physical nature (i.e. the file system becomes corrupted) are non-fatal events - there are things a system administrator can do before issuing the last rites and restoring from one of their copious backups :) You will be informed of the fact that a file system is corrupted by a harmless, but feared little messages at boot time, something like:
Cant mount /dev/hda1

If you are lucky, the system will ignore the file system problems and try to mount the corrupted partition READ ONLY. It is at this point that most people enter a hyperactive frenzy of swearing, violent screaming tantrums and self-destructive cranial impact diversions (head butting the wall).

What to do
It is important to establish that the problem is logical, not physical. There is little you can do if a disk head has crashed (on the therapeutic side, taking the offending hard disk into the car park and beating it with a stick can produce favourable results). A logical crash is something that is caused by the file system becoming confused. Things like: Many files using the one data block. Blocks marked as free but being used and vice versa. Incorrect link counts on I-Nodes. Differences in the "size of file" field in the I-Node and the number of data blocks actually used. Illegal blocks within files. I-Nodes contain information but are not in any directory entry (these type of files, when recovered, are placed in the lost+found directory). Directory entries that point to illegal or unallocated I-Nodes.

are the product of file system confusion. These problems will be detected and (usually) fixed by a program called fsck.

fsck
fsck is actually run at boot time on most Linux systems. Every x number of boots, fsck will do a comprehensive file system check. In most cases, these boot time runs of fsck automatically fix problems - though occasionally you may be prompted to confirm some fsck action. If however, fsck reports

David Jones (20.01.00)

Page 295

85321, Systems Administration

Chapter 10: Managing File Systems

some drastic problem at boot time, you will usually be thrown in to the root account and issued a message like:
************************************** fsck returned error code - REBOOT NOW! **************************************

It is probably a good idea to manually run fsck on the offending device at this point (we will get onto how in a minute). At worst, you will get a message saying that the system cant mount the file system at all and you have to reboot. It is at this point you should drag out your rescue disks (which of course contain a copy of fsck) and reboot using them. The reason for booting from an alternate source (with its own file system) is because it is quite possible that the location of the fsck program (/sbin) has become corrupted as has the fsck binary itself! It is also a good idea to run fsck only on unmounted file systems.

Using fsck
fsck is run by issuing the command:
fsck file_system

where file_system is a device or directory from which a device is mounted. fsck will do a check on all I-Nodes, blocks and directory entries. If it encounters a problem to be fixed, it will prompt you with a message. If the message asks if fsck can SALVAGE, FIX, CONTINUE, RECONNECT or ADJUST, then it is usually safe to let it. Requests involving REMOVE and CLEAR should be treated with more caution.

What caused the problem?

Problems with the file system are caused by: People turning off the power on a machine without going through the shutdown process - this is because Linux uses a very smart READ and WRITE disk cache - this cache is only flushed (or written to disk) periodically and on shutdown. fsck will usually fix these problems at the next boot. Program crashes - problems usually occur when a program is using several files and suddenly crashes without closing them. fsck usually easily fixes these problems. Kernel and system crashes - the kernel may become unstable (especially if you are using new, experimental kernels) and crash the system. Depending on the circumstances, the file system will usually be recoverable.

Exercises
10.9.

Mount the disk created in an earlier exercise. Copy the contents of your home directory to the disk. Now copy the kernel to it (/vmlinuz) but during the copy eject the disk (the idea is to do this while the light which indicates writing to disk is on). Now run fsck on that disk.
Page 296

David Jones (20.01.00)

85321, Systems Administration

Chapter 10: Managing File Systems

Conclusion
Having read and absorbed this chapter you will be aware that: Linux supports many file systems. The process of using many file systems, partitions and devices acting in concert to produce a directory structure allows for greater flexibility, performance and system integrity. The implementation of this process requires a number of components working in conjunction including: device drivers, device files, the virtual file system, specific file systems and user commands.

Review questions
10.1 As a System Administrator, you have been asked to set up a new system. The system will contain two hard disks, each 2.5 Gb in size. What issues must you consider when installing these disks? What questions should you be asking about the usage of the disks? 10.2 You have noticed that at boot time, not all the normal messages are appearing on the screen. You have also discovered that X-Windows wont run. Suggest possible reasons for this and the solutions to the problems. 10.3 A new hard disk has been added to your system to store the print spool in. List all the steps in adding this hard disk to the system. 10.4 You have just dropped your Linux box while it was running (power was lost during the systems short flight) - the system boots but will not mount the hard disk. Discuss possible reasons for the problem and the solutions. 10.5 What are links used for? What are the differences between hard and soft links?

David Jones (20.01.00)

Page 297

85321, Systems Administration

Chapter 11: Backups

Chapter
Backups
Like most of those who study history, he (Napoleon III) learned from the mistakes of the past how to make new ones. A.J.P. Taylor.

Introduction
This is THE MOST IMPORTANT responsibility of the System Administrator. Backups MUST be made of all the data on the system. It is inevitable that equipment will fail and that users will "accidentally" delete files. There should be a safety net so that important information can be recovered.

It isnt just users who accidentally delete files

A friend of mine who was once the administrator of a UNIX machine (and shall remain nameless, but is now a respected Academic at CQU) committed one of the great no-nos of UNIX Administration. Early on in his career he was carefully removing numerous old files for some obscure reason when he entered commands resembling the following (he was logged in as root when doing this).
cd / usr/user/panea rm -r * notice the mistake

The first command contained a typing mistake (the extra space) that meant that instead of being in the directory /usr/user/panea he was now in the / directory. The second command says delete everything in the current directory and any directories below it. Result: a great many files removed. The moral of this story is that everyone makes mistakes. Root users, normal users, hardware and software all make mistakes, break down or have faults. This means you must keep backups of any system.

Other Resources
Other resources which discuss backups and related information include How-tos Linux ADSM Mini-Howto, The LAME guides chapter on backup and restore procedures The Linux Systems Administrators Guides chapter (10) on backups

David Jones (20.01.00)

Page 298

85321, Systems Administration

Chapter 11: Backups

Backups arent enough

Making sure that backups are made at your site isnt enough. Backups arent any good if you cant restore the information contained. You must have some sort of plan to recover the data. That plan should take into account all sorts of scenarios. Recovery planning is not covered to any great extent in this text. That doesnt mean it isnt important.

Characteristics of a good backup strategy

Backup strategies change from site to site. What works on one machine may not be possible on another. There is no standard backup strategy. There are however a number of characteristics that need to be considered including ease of use, time efficiency, ease of restoring files, ability to verify backups, tolerance of faulty media, and portabilty to a range of machines.

Ease of use
If backups are easy to use, you will use them. AUTOMATE!! It should be as easy as placing a tape in a drive, typing a command and waiting for it to complete. In fact you probably shouldnt have to enter the command, it should be automatically run. When backups are too much work At many large computing sites operators are employed to perform low-level tasks like looking after backups. Looking after backups generally involves obtaining a blank tape, labelling it, placing it in the tape drive, waiting for the information to be stored on the tape and then storing it away. A true story that is told by an experienced Systems Administrator is about an operator that thought backups took too long to perform. To solve this problem the operator decided backups finished much quicker if you didnt bother putting the tape in the tape drive. You just labelled the blank tape and placed it in storage. Quite alright as long as you dont want to retrieve anything from the backups.

David Jones (20.01.00)

Page 299

85321, Systems Administration

Chapter 11: Backups

Time efficiency
Obtain a balance to minimise the amount of operator, real and CPU time taken to carry out the backup and to restore files. The typical tradeoff is that a quick backup implies a longer time to restore files. Keep in mind that you will in general perform more backups than restores. On some large sites, particular backup strategies fail because there arent enough hours in a day. Backups scheduled to occur every 24 hours fail because the previous backup still hasn't finished. This obviously occurs at sites which have large disks.

Ease of restoring files

The reason for doing backups is so you can get information back. You will have to be able to restore information ranging from a single file to an entire file system. You need to know on which media the required file is and you need to be able to get to it quickly. This means that you will need to maintain a table of contents and label media carefully.

Ability to verify backups

YOU MUST VERIFY YOUR BACKUPS. The safest method is once the backup is complete, read the information back from the media and compare it with the information stored on the disk. If it isnt the same then the backup is not correct. Well that is a nice theory but it rarely works in practice. This method is only valid if the information on the disk hasn't changed since the backup started. This means the file system cannot be used by users while a backup is being performed or during the verification. Keeping a file system unused for this amount of time is not often an option. Other quicker methods include restoring a random selection of files from the start, middle and end of the backup, If these particular files are retrieved correctly the assumption is that all of the files are valid. create a table of contents during the backup; afterwards read the contents of the tape and compare the two.

These methods also do not always work. Under some conditions and with some commands the two methods will not guarantee that your backup is correct.

Tolerance of faulty media

A backup strategy should be able to handle faults in the media, and
Page 300

David Jones (20.01.00)

85321, Systems Administration

Chapter 11: Backups

physical dangers. there exist at least two copies of full backups of a system, and that at least one set should be stored at another site.

There are situations where it is important that

Consider the following situation. A site has one set of full backups stored on tapes. They are currently performing another full backup of the system onto the same tapes. What happens when the backup system is happily churning away when it gets about halfway and crashes (the power goes off, the tape drive fails etc). This could result in the both the tape and the disk drive being corrupted. Always maintain duplicate copies of full backups. An example of the importance of storing backups off site was the Pauls icecream factory in Brisbane. The factory is located right on the riverbank and during the early 1970s Brisbane suffered problems caused by a major flood. The Pauls computer room was in the basement of their factory and was completely washed out. All the backups were kept in the computer room.

Portabilty to a range of platforms

There may be situations where the data stored on backups must be retrieved onto a different type of machine. The ability for backups to be portable to different types of machine is often an important characteristic. For example: The computer currently being used by a company is the last in its line. The manufacturer is bankrupt and no one else uses the machine. Due to unforeseen circumstances the machine burns to the ground. The Systems Administrator has recent backups available and they contain essential data for this business. How are the backups to be used to reconstruct the system?

Considerations for a backup strategy

Apart from the above characteristics, factors that may affect the type of backup strategy implemented will include the available commands The characteristics of the available commands limit what can be done. available hardware The capacity of the backup media to be used also limits how backups are performed. In particular how much information can the media hold? maximum expected size of file systems The amount of information required to be backed up and whether or not the combination of the available software and hardware can handle it. A suggestion is that individual file systems should never contain more information than can fit easily onto the backup media.

David Jones (20.01.00)

Page 301

85321, Systems Administration

Chapter 11: Backups

importance of the data The more important the data is, the more important that it be backed up regularly and safely. level of data modification The more data being created and modified, the more often it should be backed up. For example the directories /bin and /usr/bin will hardly ever change so they rarely need backing up. On the other hand directories under /home are likely to change drastically every day.

The components of backups

There are basically three components to a backup strategy. The scheduler Decides when the backup is performed. transport, and The command that moves the backup from the disks to the backup media. media The actual physical device on which the backup is stored.

Scheduler
The scheduler is the component that decides when backups should be performed and how much should be backed up. The scheduler could be the root user or a program, usually cron (discussed in a later chapter). The amount of information that the scheduler backs up can have the following categories full backups, All the information on the entire system is backed up. This is the safest type but also the most expensive in machine and operator time and the amount of media required. partial backups, or Only the busier and more important file systems are backed up. One example of a partial backup might include configuration files (like /etc/passwd), user home directories and the mail and news spool directories. The reasoning is that these files change the most and are the most important to keep a track of. In most instances this can still take substantial resources to perform. incremental backups. Only those files that have been modified since the last backup are backed up. This method requires less resources but a large amount of incremental backups make it more difficult to locate the version of a particular file you may desire.

David Jones (20.01.00)

Page 302

85321, Systems Administration

Chapter 11: Backups

Transport
The transport is a program that is responsible for placing the backed-up data onto the media. There are quite a number of different programs that can be used as transports. Some of the standard UNIX transport programs are examined later in this chapter. There are two basic mechanisms that are used by transport programs to obtain the information from the disk image, and through the file system.

Image transports An image transport program bypasses the file system and reads the information straight off the disk using the raw device file. To do, this the transport program needs to understand how the information is structured on the disk. This means that transport programs are linked very closely to exact file systems since different file systems structure information differently. Once read off the disk, the data is written byte by byte from disk onto tape. This method generally means that backups are usually quicker than the "file by file" method. However restoration of individual files generally takes much more time. Transport programs that use the method include dd, volcopy and dump. File by file Commands performing backups using this method use the system calls provided by the operating system to read the information. Since almost any UNIX system uses the same system calls, a transport program that uses the file by file method (and the data it saves) is more portable. File by file backups generally take more time but it is generally easier to restore individual files. Commands that use this method include tar and cpio. Backing up FAT and EXT2 file systems If you are like most people using this text then chances are that your Linux computer contains both FAT and EXT2 file systems. The FAT file systems will be used by the version of Windows you were originally running while the EXT2 file systems will be those used by Linux. Of course being the trainee computing professional you are backups of your personal computer are performed regularly. It would probably be useful to you to be able to backup both the FAT and EXT2 file systems at the same time, without having to switch operating systems.

David Jones (20.01.00)

Page 303

85321, Systems Administration

Chapter 11: Backups

Well doing this from Windows isnt going to work. Windows still doesnt read the EXT2 file system. So you will have to do it from Linux. Which type of transport do you use for this, image or file by file? Well heres a little excerpt from the man page for the dump command, the one of the image transports available on Linux.
It might be considered a bug that this version of dump can only handle ext2 filesystems. Specifically, it does not work with FAT filesystems.

If you think about it this shortcoming is kind of obvious. The dump command does not use the kernel file system code. It is an image transport. This means it must know everything about the filesystem it is going to backup. How are directories structured, how are the data blocks for files store on the system, how is file metadata (e.g. permissions, file owners etc) stored and many more questions. The people who wrote dump included this information into the command. They didnt include any information about the FAT file system. So dump cant backup FAT file systems. File by file transports on the other hand can quite happily backup any file system which you can mount on a Linux machine. In this situation the virtual file system takes care of all the differences and all the file-by-file transport knows about are what appear to be normal Linux files.

Media
Backups are usually made to tape based media. There are different types of tape. Tape media can differ in physical size and shape, and amount of information that can be stored. From 100Mb up to 8Gb.

Different types of media can also be more reliable and efficient. The most common type of backup media used today are 4 millimetre DAT tapes. Reading Under the Resource Materials section for Week 6 on the 85321 Web site/CD-ROM you will find a pointer to the USAIL resources on backups. This includes a pointer to discussion about the different type of media which are available.

Commands
As with most things, the different versions of UNIX provide a plethora of commands that could possibly act as the transport in a backup system. The following table provides a summary of the characteristics of the more common programs that are used for this purpose.

David Jones (20.01.00)

Page 304

85321, Systems Administration

Chapter 11: Backups

Command

Availability

Characteristics image backup, allows multiple volumes, not included on most AT&T systems file by file, most versions do not support multiple volumes, intolerant of errors file by file, can support multiple volumes some versions dont,

dump/restore BSD systems tar cpio almost all systems AT&T systems

The

Table 11.1. Different Backup Commands.

There are a number of other public domain and commercial backup utilities available which are not listed here.

dump and restore

A favourite amongst many Systems Administrators, dump is used to perform backups and restore is used to retrieve information from the backups. These programs are of BSD UNIX origin and have not made the jump across to SysV systems. Most SysV systems do not come with dump and restore. The main reason is that since dump and restore bypass the file system, they must know how the particular file system is structured. So you simply cant recompile a version of dump from one machine onto another (unless they use the same file system structure). Many recent versions of systems based on SVR4 (the latest version of System V UNIX) come with versions of dump and restore. dump on Linux There is a version of dump for Linux. However, it may be possible that you do not have it installed on your system. RedHat Linux does include an RPM package which contains dump. If your system doesnt have dump and restore installed you should install it now. RedHat provides a couple of tools to installe these packages: rpm and glint. glint is the GUI tool for managing packages. Refer to the RedHat documentation for more details on using these tools. dump The command line format for dump is dump [ options [ arguments ] ] file system dump [ options [ arguments ] ] filename Arguments must appear after all options and must appear in a set order. dump is generally used to backup an entire partition (file system). If given a list of filenames, dump will backup the individual files.

David Jones (20.01.00)

Page 305

85321, Systems Administration

Chapter 11: Backups

dump works on the concept of levels (it uses 9 levels). A dump level of 0 means that all files will be backed up. A dump level of 1...9 means that all files that have changed since the last dump of a lower level will be backed up. Table 11.2 shows the arguments for dump. Options 0-9 dump level Purpose

a archive-file archive-file will be a table of contents of the archive. f dump-file u v specify the file (usually a device file) to write the dump to, a specifies standard output update the dump record (/etc/dumpdates) after writing each volume, rewind the tape and verify. The file system must not be used during dump or the verification.
Table 11.2. Arguments for dump

There are other options. Refer to the man page for the system for more information. For example:
dump 0dsbfu 54000 6000 126 /dev/rst2 /usr

full backup of /usr file system on a 2.3 Gig 8mm tape connected to device rst2 The numbers here are special information about the tape drive the backup is being written on. The restore command The purpose of the restore command is to extract files archived using the dump command. restore provides the ability to extract single individual files, directories and their contents and even an entire file system.
restore -irRtx [ modifiers ] [ filenames ]

The restore command has an interactive mode where commands like ls etc can be used to search through the backup.

David Jones (20.01.00)

Page 306

85321, Systems Administration

Chapter 11: Backups

Arguments i

Purpose interactive, directory information is read from the tape after which you can browse through the directory hierarchy and select files to be extracted. restore the entire tape. Should only be used to restore an entire file system or to restore an incremental tape after a full level 0 restore. table of contents, if no filename provided, root directory is listed including all subdirectories (unless the h modifier is in effect) extract named files. If a directory is specified, it and all its subdirectories are extracted.
Table 11.3. Arguments for the restore Command.

t x

Modifiers

Purpose

a archive-file use an archive file to search for a files location. Convert contents of the dump tape to the new file system format d h v f dump-file s n turn on debugging prevent hierarchical restoration of subdirectories verbose mode specify dump-file to use, - refers to standard input skip to the nth dump file on the tape
Command.

Table 11.4. Argument modifiers for the restore

Using dump and restore without a tape

Not many of you will have tape drives or similar backup media connected to your Linux machine. However, it is important that you experiment with the dump and restore commands to gain an understanding of how they work. This section offers a little kludge which will allow you to use these commands without a tape drive. The method relies on the fact that UNIX accesses devices through files.

Our practice file system

For all our experimentation with the commands in this chapter we are going to work with a practice file system. Practising backups with hard-drive partitions
David Jones (20.01.00) Page 307

85321, Systems Administration

Chapter 11: Backups

is not going to be all that efficient as they will almost certainly be very large. Instead we are going to work with a floppy drive. The first step then is to format a floppy with the ext2 file system. By now you should know how to do this. Heres what I did to format a floppy and put some material on it.
[root@beldin]# /sbin/mke2fs /dev/fd0 mke2fs 1.10, 24-Apr-97 for EXT2 FS 0.5b, 95/08/09 Linux ext2 filesystem format Filesystem label= 360 inodes, 1440 blocks 72 blocks (5.00%) reserved for the super user First data block=1 Block size=1024 (log=0) Fragment size=1024 (log=0) 1 block group 8192 blocks per group, 8192 fragments per group 360 inodes per group Writing inode tables: done Writing superblocks and filesystem accounting information: done [root@beldin]# mount -t ext2 /dev/fd0 /mnt/floppy [root@beldin]# cp /etc/passwd /etc/issue /etc/group /var/log/messages /mnt/floppy [root@beldin dump-0.3]#

Doing a level 0 dump

So Ive copied some important stuff to this disk. Lets assume I want to do a level 0 dump of the /mnt/floppy file system. How do I do it?
[root@beldin]# /sbin/dump 0f /tmp/backup /mnt/floppy DUMP: Date of this level 0 dump: Sun Jan 25 15:05:11 1998 DUMP: Date of last level 0 dump: the epoch DUMP: Dumping /dev/fd0 (/mnt/floppy) to /tmp/backup DUMP: mapping (Pass I) [regular files] DUMP: mapping (Pass II) [directories] DUMP: estimated 42 tape blocks on 0.00 tape(s). DUMP: dumping (Pass III) [directories] DUMP: dumping (Pass IV) [regular files] DUMP: DUMP: 29 tape blocks on 1 volumes(s) DUMP: Closing /tmp/backup DUMP: DUMP IS DONE

The arguments to the dump command are 0 This tells dump I wish to perform a level 0 dump of the file system. f This is telling dump that I will tell it the name of the file that it should write the backup to. /tmp/backup This is the name of the file I want the backup to go to. Normally, this would be the device file for a tape drive or other backup device. However, since I dont have one Im telling it a normal file. /mnt/floppy This is the file system I want to backup.
Page 308

David Jones (20.01.00)

85321, Systems Administration

Chapter 11: Backups

What this means is that I have now created a file, /tmp/backup, which contains a level 0 dump of the floppy.
[root@beldin]# ls -l /tmp/backup -rw-rw-r-1 root tty 20480 Jan 25 15:05 /tmp/backup

Restoring the backup

Now that we have a dump archive to work with, we can try using the restore command to retrieve files.
[root@beldin dump-0.3]# /sbin/restore -if /tmp/backup
restore > ? Available commands are: ls [arg] - list directory cd arg - change directory pwd - print current directory add [arg] - add arg to list of files to be extracted delete [arg] - delete arg from list of files to be extracted extract - extract requested files setmodes - set modes of requested directories quit - immediately exit program what - list dump header information verbose - toggle verbose flag (useful with ls) help or ? - print this list If no arg is supplied, the current directory is used restore > ls .: group issue lost+found/ messages passwd restore > add passwd restore > extract You have not read any tapes yet. Unless you know which volume your file(s) are on you should start with the last volume and work towards towards the first. Specify next volume #: 1 Mount tape volume 1 Enter none if there are no more tapes otherwise enter tape name (default: /tmp/backup) set owner/mode for .? [yn] y restore > quit [root@beldin]# ls -l passwd -rw-r--r-1 root root 787 Jan 25 15:00 passwd

Alternative
Rather than backup to a normal file on the hard-drive you could choose to backup files directly to a floppy drive (i.e. use /dev/fd0 rather than /tmp/backup). One problem with this alternative is that you are limited to 1.44Mb per media. This used to be a problem because the Linux version of dump did not support multiple volumes. It appears it know does. Exercises
11.1.

Do a level 0 dump of a portion of your home directory onto a file somewhere on your hard drive. Examine the file /etc/dumpdates. How has it changed? Use restore to retrieve some individual files from the backup and also to retrieve the entire backup.

11.2.

David Jones (20.01.00)

Page 309

85321, Systems Administration

Chapter 11: Backups

11.3.

Perform a dump onto floppies which requires more than 1.44Mb of space (the idea here is to play around with multiple volume backups). You can do this by creating a directory and placing files into it until you have more than 1.44Mb of data in it (use the du command to find out how much space is being consumed). After youve backed up onto floppies try retrieving some files. What problems do you face?

The tar command

tar is a general purpose command used for archiving files. It takes multiple files and directories and combines them into one large file. By default the resulting file is written to a default device (usually a tape drive). However the resulting file can be placed onto a disk drive.
tar -function[modifier] device [files]

The purpose and values for function and modifier are shown in Tables 11.5 through 11.7. When using tar, each individual file stored in the final archive is preceded by a header that contains approximately 512 bytes of information. Also the end of the file is always padded so that it occurs on an even block boundary. For this reason, every file added into the tape archive has on average an extra .75Kb of padding per file. Arguments function modifier files Purpose A single letter specifying what should be done, values listed in Table 11.6 Letters that modify the action of the specified function, values listed in Table 11.7 The names of the files and directories to be restored or archived. If it is a directory then EVERYTHING in that directory is restored or archived
Table 11.5. Arguments to tar.

David Jones (20.01.00)

Page 310

85321, Systems Administration

Chapter 11: Backups

Function c r t u * x

Purpose create a new tape, do not write after last file replace, the named files are written onto the end of the tape table, information about specified files is listed, similar in output to the command ls -l, if no files specified all files listed update, named files are added to the tape if they are not already there or they have been modified since being previously written extract, named files restored from the tape, if the named file matches a directory all the contents are extracted recursively
Values of * the u function can be very slow Table 11.6. the function argument for tar.

Modifier v w f m

Purpose verbose, tar reports what it is doing and to what tar prints the action to be taken, the name of the file and waits for user confirmation file, causes the device parameter to be treated as a file modify, tells tar not to restore the modification times as they were archived but instead to use the time of extraction ownership, use the UID and GID of the user running tar not those stored on the tape
Values of the Table 11.7. modifier argument for tar.

If the f modifier is used it must be the last modifier used. Also tar is an example of a UNIX command where the - character is not required to specify modifiers. For example:
tar -xvf temp.tar tar xvf temp.tar

extracts all the contents of the tar file temp.tar

tar -xf temp.tar hello.dat

extracts the file hello.dat from the tar file temp.tar

tar -cv /dev/rmt0 /home

archives all the contents of the /home directory onto tape, overwriting whatever is there

David Jones (20.01.00)

Page 311

85321, Systems Administration

Chapter 11: Backups

Exercises
11.4.

Create a file called temp.dat under a directory tmp that is within your home directory. Use tar to create an archive containing the contents of your home directory. Delete the $HOME/tmp/temp.dat created in the previous question. Extract the copy of the file that is stored in the tape archive (the term tape archive is used to refer to a file created by tar) created in the previous question.

11.5.

The dd command
The man page for dd lists its purpose as being "copy and convert data". Basically dd takes input from one source and sends it to a different destination. The source and destination can be device files for disk and tape drives, or normal files. The basic format of dd is
dd [option = value ....]

Table 11.8. lists some of the different options available. Option if=name of=name ibs=num obs=num bs=num skip=num files=num conv=ascii conv=ebcdic conv=lcase conv=ucase conv=swab Purpose input file name (default is standard input) output file name (default is standard output) the input block size in num bytes (default is 512) the output block size in num bytes (default is 512) set both input and output block size skip num input records before starting to copy copy num files before stopping (used when input is from magnetic tape) convert EBCDIC to ASCII convert ASCII to EBCDIC make all letters lowercase make all letters uppercase swap every pair of bytes
Table 11.8. Options for dd.

David Jones (20.01.00)

Page 312

85321, Systems Administration

Chapter 11: Backups

For example:
dd if=/dev/hda1 of=/dev/rmt4

with all the default settings copy the contents of hda1 (the first partition on the first disk) to the tape drive for the system Exercises
11.6.

Use dd to copy the contents of a floppy disk to a single file to be stored under your home directory. Then copy it to another disk.

The mt command
The usual media used in backups is magnetic tape. Magnetic tape is a sequential media. That means that to access a particular file you must pass over all the tape containing files that come before the file you want. The mt command is used to send commands to a magnetic tape drive that control the location of the read/write head of the drive.
mt [-f tapename] command [count]

Arguments tapename command

Purpose raw device name of the tape device one of the commands specified in table 11.10. Not all commands are recognised by all tape drives. number of times to carry out command

count

Table 11.9. Parameters for the mt Command.

Commands fsf asf rewind retension erase offline

Action move forward the number of files specified by the count argument move forward to file number count rewind the tape wind the tape out to the end and then rewind erase the entire tape eject the tape
mt Command.

Table 11.10. Commands Possible using the

For example:
mt -f /dev/nrst0 asf 3

moves to the third file on the tape

David Jones (20.01.00)

Page 313

85321, Systems Administration

Chapter 11: Backups

mt -f /dev/nrst0 rewind mt -f /dev/nrst0 fsf 3

same as the first command The mt command can be used to put multiple dump/tar archive files onto the one tape. Each time dump/tar is used, one file is written to the tape. The mt command can be used to move the read/write head of the tape drive to the end of that file, at which time dump/tar can be used to add another file. For example:
mt -f /dev/rmt/4 rewind

rewinds the tape drive to the start of the tape

tar -cvf /dev/rmt/4 /home/jonesd

backs up my home directory, after this command the tape will be automatically rewound
mt -f /dev/rmt/4 asf 1

moves the read/write head forward to the end of the first file
tar -cvf /dev/rmt/4a /home/thorleym

backs up the home directory of thorleym onto the end of the tape drive There are now two tar files on the tape, the first containing all the files and directories from the directory /home/jonesd and the second containing all the files and directories from the directory /home/thorleym.

Compression programs
Compression programs are sometimes used in conjunction with transport programs to reduce the size of backups. This is not always a good idea. Adding compression to a backup adds extra complexity to the backup and as such increases the chances of something going wrong. compress compress is the standard UNIX compression program and is found on every UNIX machine (well, I dont know of one that doesnt have it). The basic format of the compress command is
compress filename

The file with the name filename will be replaced with a file with the same name but with an extension of .Z added, and that is smaller than the original (it has been compressed). A compressed file is uncompressed using the uncompress command or the -d switch of compress.
uncompress filename

or

compress -d filename

David Jones (20.01.00)

Page 314

85321, Systems Administration

Chapter 11: Backups

For example:
bash$ ls -l ext349* -rw-r----- 1 jonesd 17340 Jul 16 14:28 ext349

bash$ compress ext349 bash$ ls -l ext349* -rw-r----- 1 jonesd bash$ uncompress ext349 bash$ ls -l ext349* -rw-r----- 1 jonesd

5572 Jul 16 14:28 ext349.Z

17340 Jul 16 14:28 ext349

gzip
gzip is a new addition to the UNIX compression family. It works in basically the same way as compress but uses a different (and better) compression algorithm. It uses an extension of .z and the program to uncompress a gzip archive is gunzip. For example:
bash$ gzip ext349 bash$ ls -l ext349*

-rw-r----- 1 jonesd
bash$ gunzip ext349

4029 Jul 16 14:28 ext349.z

Exercises
11.7.

Modify your solution to exercise 11.5 so that instead of writing the contents of your floppy straight to a file on your hard disk it first compresses the file using either compress or gzip and then saves to a file.

Conclusions
In this chapter you have been introduced to the components of a backup strategy scheduler, transport, and media been shown some of the UNIX commands that can be used as the transport in a backup strategy examined some of the characteristics of a good backup strategy and some of the factors that affect a backup strategy

Review questions
11.1. Design a backup strategy for your system. List the components of your backup strategy and explain how these components affect your backup strategy.

David Jones (20.01.00)

Page 315

85321, Systems Administration

Chapter 11: Backups

11.2. Explain the terms media, scheduler and transport. 11.3. Outline the difference between file by file and image transport programs. 11.4. ACME Backup Systems has just produced a wonderful backup system which has caught the eye of your manager. He has decided that it is the product you should be using to backup the Linux servers within your organization. ACMEs Backup-true product is an image transport developed for the Windows NT file system. Why cant you use it for to backup your Linux systems?

David Jones (20.01.00)

Page 316

85321, Systems Administration

Chapter 12: Startup and Shutdown

Chapter
Startup and Shutdown

Introduction
Being a multi-tasking, multi-user operating system means that UNIX is a great deal more complex than an operating system like MS-DOS. Before the UNIX operating system can perform correctly, there are a number of steps that must be followed, and procedures executed. The failure of any one of these can mean that the system will not start, or if it does it will not work correctly. It is important for the Systems Administrator to be aware of what happens during system startup so that any problems that occur can be remedied. It is also important for the Systems Administrator to understand what the correct mechanism is to shut a UNIX machine down. A UNIX machine should (almost) never be just turned off. There are a number of steps to carry out to ensure that the operating system and many of its support functions remain in a consistent state. By the end of this chapter you should be familiar with the startup and shutdown procedures for a UNIX machine and all the related concepts.

Other Resources
There is a lot of available information about the startup process of a Linux machine and also how you recover from errors in the startup process. These include HOW-TOs BootPrompt HOW-TO, Boot disk HOW-TO, UPS HOW-TO, LILO Mini HOW-TO, Win95 + WinNT + Linux multiboot using LILO mini-HOWTO Rescue disk sets The Linux Systems Administrators Guide (part of the LDP and on the 85321 CD-ROM) chapter (chapter 6) on boots and shutdowns and also Chapter 7 on init.

A booting overview
The process by which a computer is turned on and the UNIX operating system starts functioning booting - consists of the following steps finding the kernel, The first step is to find the kernel of the operating system. How this is achieved is usually particular to the type of hardware used by the computer.

David Jones (20.01.00)

Page 317

85321, Systems Administration

Chapter 12: Startup and Shutdown

starting the kernel, In this step the kernel starts operation and in particular goes looking for all the hardware devices that are connected to the machine. starting the processes. All the work performed by a UNIX computer is done by processes. In this stage, most of the system processes and daemons are started. This step also includes a number of steps which configure various services necessary for the system to work.

Finding the Kernel

For a UNIX computer to be functional it must have a kernel. The kernel provides a number of essential services which are required by the rest of the system in order for it to be functional. This means that the first step in the booting process of a UNIX computer is finding out where the kernel is. Once found, it can be started, but thats the next section.

ROM
Most machines have a section of read only memory (ROM) that contains a program the machine executes when the power first comes on. What is programmed into ROM will depend on the hardware platform. For example, on an IBM PC, the ROM program typically does some hardware probing and then looks in a number of predefined locations (the first floppy drive and the primary hard drive partition) for a bootstrap program. On hardware designed specifically for the UNIX operating system (machines from DEC, SUN etc), the ROM program will be a little more complex. Many will present some form of prompt. Generally this prompt will accept a number of commands that allow the Systems Administrator to specify where to boot the machine from, and Sometimes the standard root partition will be corrupt and the system will have to be booted from another device. Examples include another hard drive, a CD-ROM, floppy disk or even a tape drive. whether to come up in single user or multi-user mode. As a bare minimum, the ROM program must be smart enough to work out where the bootstrap program is stored and how to start executing it. The ROM program generally doesnt know enough to know where the kernel is or what to do with it.

The bootstrap program

At some stage the ROM program will execute the code stored in the boot block of a device (typically a hard disk drive). The code stored in the boot block is referred to as a bootstrap program. Typically the boot block isnt big enough to hold the kernel of an operating system so this intermediate stage is necessary. The bootstrap program is responsible for locating and loading (starting) the kernel of the UNIX operating system into memory. The kernel of a UNIX
David Jones (20.01.00) Page 318

85321, Systems Administration

Chapter 12: Startup and Shutdown

operating system is usually stored in the root directory of the root file system under some system-defined filename. Newer versions of Linux, including RedHat 5.0 and above, put the kernel into a directory called /boot. /boot is often on a separate partition. In fact the default installation of RedHat Linux will create /boot as a separate partition. The most common bootstrap program in the Linux world is a program called LILO. Reading LILO is such an important program to the Linux operating system that it has its own HOW-TO. The HOW-TO provides a great deal of information about the boot process of a Linux computer.

Booting on a PC
The BIOS on a PC generally looks for a bootstrap program in one of two places (usually in this order) the first (A:) floppy drive, or the first (C:) hard drive.

By playing with your BIOS settings you can change this order or even prevent the BIOS from checking one or the other. The BIOS loads the program that is on the first sector of the chosen drive and loads it into memory. This bootstrap program then takes over. For example, making sure people cant boot your Linux machine of a floppy can prevent them from gaining access to the data on your machine.

On the floppy
On a bootable floppy disk the bootstrap program simply knows to load the first blocks on the floppy that contain the kernel into a specific location in memory. A normal Linux boot floppy contains no file system. It simply contains the kernel copied into the first sectors of the disk. The first sector on the disk contains the first part of the kernel which knows how to load the remainder of the kernel into RAM. This means you cant mount the boot floppy onto your Linux machine and read the contents of the disk using ls and other associated commands.

Making a boot disk

The simplest method for creating a floppy disk which will enable you to boot a Linux computer is insert a floppy disk into a computer already running Linux login as root change into the /boot directory

David Jones (20.01.00)

Page 319

85321, Systems Administration

Chapter 12: Startup and Shutdown

copy the current kernel onto the floppy

dd if=vmlinuz of=/dev/fd0

The name of the kernel, vmlinuz, may change from system to system. For example, on some version 5.0 of RedHat Linux it will have been vmlinux-2.0.31. Exercises
12.1.

Using the above steps create a boot floppy for your machine and test it out.

Using a boot loader

Having a boot floppy for your system is a good idea. It can come in handy if you do something to your system which prevents the normal boot procedure from working. One example of this is when you are compiling a new kernel. It is not unheard of for people to create a kernel which will not boot their system. If you dont have an alternative boot method in this situation then you will have some troubles. However, you cant use this process to boot from a hard-drive. Instead a boot loader or boot strap program, such as LILO, is used. A boot loader generally examines the partition table of the hard-drive, identifies the active partition, and then reads and starts the code in the boot sector for that partition. This is a simplification. In reality the boot loader must identify, somehow, the sectors in which the kernel resides. Other features a boot loader (under Linux) offers include using a key press to bring up a prompt to modify the boot procedure, and the passing of parameters to the kernel to modify its operation

Exercises
12.2.

If you have the time, havent done so already, or know it is destined to failure read the LILO documentation and install LILO onto your system. There are some situations where you SHOULD NOT install LILO. These are outlined in the documentation. Make sure you take notice of these situations.

Starting the kernel

Okay, the boot strap program or the ROM program has found your systems kernel. What happens during the startup process? The kernel will go through the following process initialise its internal data structures, Things like ready queues, process control blocks and other data structures need to be readied. check for the hardware connected to your system, It is important that you are aware that the kernel will only look for

David Jones (20.01.00)

Page 320

85321, Systems Administration

Chapter 12: Startup and Shutdown

hardware that it contains code for. If your system has a SCSI disk drive interface your kernel must have the SCSI interface code before it will be able to use it. verify the integrity of the root file system and then mount it, and create the process 0 (swapper) and process 1 (init).

The swapper process is actually part of the kernel and is not a "real" process. The init process is the ultimate parent of all processes that will execute on a UNIX system. Once the kernel has initialised itself, init will perform the remainder of the startup procedure.

Kernel boot messages

When a UNIX kernel is booting, it will display messages on the main console about what it is doing. Under Linux, these messages are also sent to the file /var/log/dmesg. The following is a copy of the boot messages on my machine. Examine the messages that your kernel displays during bootup and compare them with mine.
Linux version 2.2.12-20 ([email protected]) (gcc version egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)) #1 Mon Sep 27 10:25:54 EDT 1999 Detected 233867806 Hz processor. Console: colour VGA+ 80x25 Calibrating delay loop... 466.94 BogoMIPS Memory: 127668k/131072k available (1008k kernel code, 412k reserved, 1640k data, 64k init) DENTRY hash table entries: 262144 (order: 9, 2097152 bytes) Buffer-cache hash table entries: 131072 (order: 7, 524288 bytes) Page-cache hash table entries: 32768 (order: 5, 131072 bytes) VFS: Diskquotas version dquot_6.4.0 initialized CPU: Intel Mobile Pentium MMX stepping 01 Checking 386/387 coupling... OK, FPU using exception 16 error reporting. Checking hlt instruction... OK. Intel Pentium with F0 0F bug - workaround enabled. POSIX conformance testing by UNIFIX PCI: PCI BIOS revision 2.10 entry at 0xf56ee PCI: Using configuration type 1 PCI: Probing PCI hardware PCI: Enabling I/O for device 00:0a Linux NET4.0 for Linux 2.2 Based upon Swansea University Computer Society NET3.039 NET4: Unix domain sockets 1.0 for Linux NET4.0. NET4: Linux TCP/IP 1.0 for NET4.0 IP Protocols: ICMP, UDP, TCP, IGMP TCP: Hash tables configured (ehash 131072 bhash 65536) Initializing RT netlink socket Starting kswapd v 1.5 Detected PS/2 Mouse Port. Serial driver version 4.27 with MANY_PORTS MULTIPORT SHARE_IRQ enabled ttyS00 at 0x03f8 (irq = 4) is a 16550A pty: 256 Unix98 ptys configured apm: BIOS version 1.2 Flags 0x03 (Driver version 1.9) Real Time Clock Driver v1.09 RAM disk driver initialized: 16 RAM disks of 4096K size PIIX4: IDE controller on PCI bus 00 dev 09 PIIX4: not 100% native mode: will probe irqs later ide0: BM-DMA at 0xfcd0-0xfcd7, BIOS settings: hda:pio, hdb:pio ide1: BM-DMA at 0xfcd8-0xfcdf, BIOS settings: hdc:pio, hdd:pio hda: IBM-DCXA-210000, ATA DISK drive hdc: TOSHIBA CD-ROM XM-1602B, ATAPI CDROM drive ide0 at 0x1f0-0x1f7,0x3f6 on irq 14 ide1 at 0x170-0x177,0x376 on irq 15 hda: IBM-DCXA-210000, 9590MB w/420kB Cache, CHS=1222/255/63

David Jones (20.01.00)

Page 321

85321, Systems Administration

Chapter 12: Startup and Shutdown

hdc: ATAPI 20X CD-ROM drive, 128kB Cache Uniform CDROM driver Revision: 2.56 Floppy drive(s): fd0 is 1.44M FDC 0 is a National Semiconductor PC87306 md driver 0.90.0 MAX_MD_DEVS=256, MAX_REAL=12 raid5: measuring checksumming speed raid5: MMX detected, trying high-speed MMX checksum routines pII_mmx : 340.614 MB/sec p5_mmx : 405.003 MB/sec 8regs : 246.888 MB/sec 32regs : 184.785 MB/sec using fastest function: p5_mmx (405.003 MB/sec) scsi : 0 hosts. scsi : detected total. md.c: sizeof(mdp_super_t) = 4096 Partition check: hda: hda1 hda2 < hda5 hda6 > RAMDISK: Compressed image found at block 0 autodetecting RAID arrays autorun ... ... autorun DONE. VFS: Mounted root (ext2 filesystem). autodetecting RAID arrays autorun ... ... autorun DONE. VFS: Mounted root (ext2 filesystem) readonly. change_root: old root has d_count=1 Trying to unmount old root ... okay Freeing unused kernel memory: 64k freed Adding Swap: 72252k swap-space (priority -1) ad1848/cs4248 codec driver Copyright (C) by Hannu Savolainen 1993-1996 [MSS: IRQ Conflict?] ad1848: Interrupt test failed (IRQ7) YM3812 and OPL-3 driver Copyright (C) by Hannu Savolainen, Rob Hooft 1993-1996

The last few lines of this output demonstrates one of the advantages of checking the kernel boot messages. Ive just discovered that the sound configuration for my system is not working as expected. Something I need to investigate and fix.

Starting the processes

So at this stage the kernel has been loaded, it has initialised its data structures and found all the hardware devices. At this stage your system cant do anything. The operating system kernel only supplies services which are used by processes. The question is how are these other processes created and executed. On a UNIX system the only way in which a process can be created is by an existing process performing a fork operation. A fork creates a brand new process that contains copies of the code and data structures of the original process. In most cases the new process will then perform an exec that replaces the old code and data structures with that of a new program. But who starts the first process? init is the process that is the ultimate ancestor of all user processes on a UNIX system. It always has a Process ID (PID) of 1. init is started by the operating system kernel so it is the only process that doesnt have a process as a parent. init is responsible for starting all other services provided by the UNIX system. The services it starts are specified by inits configuration file, /etc/inittab.

David Jones (20.01.00)

Page 322

85321, Systems Administration

Chapter 12: Startup and Shutdown

Run levels
init is also responsible for placing the computer into one of a number of run levels. The run level a computer is in controls what services are started (or stopped) by init. Table 12.2 summarises the different run levels used by RedHat Linux. At any one time, the system must be in one of these run levels. Run level
0 1

Description Halt the machine Single user mode. All file systems mounted, only small set of kernel processes running. Only root can login. multi-user mode , without remote file sharing multi-user mode with remote file sharing, processes, and daemons user definable system state used for to start X11 on boot shutdown and reboot ondemand run levels same as single-user mode, only really used by scripts
Table 12.1 Run levels

4 5 6 a b c s

or S

When a Linux system boots, init examines the /etc/inittab file for an entry of type initdefault. This entry will determine the initial run level of the system. Under Linux, the telinit command is used to change the current run level. telinit is actually a soft link to init. telinit accepts a single character argument from the following
0 1 2 3 4 5 6

The run level is switched to this level.

Q q

Tells init that there has been a change to /etc/inittab (its configuration file) and that it should re-examine it.
S s

Tells init to switch to single user mode.

David Jones (20.01.00)

Page 323

85321, Systems Administration

Chapter 12: Startup and Shutdown

/etc/inittab
/etc/inittab is the configuration file for init. It is a colon delimited field where # characters can be used to indicate comments. Each line corresponds to

a single entry and is broken into four fields the identifier One or two characters to uniquely identify the entry. the run level Indicates the run level at which the process should be executed the action Tells init how to execute the process the process The full path of the program or shell script to execute.

What happens When init is first started it determines the current run level (by matching the entry in /etc/inittab with the action initdefault) and then proceeds to execute all of the commands of entries that match the run level. The following is an example /etc/inittab taken from a RedHat machine with some comments added. Specify the default run level
id:3:initdefault: # System initialisation. si::sysinit:/etc/rc.d/rc.sysinit

when first entering various runlevels run the related startup scripts before going any further
l0:0:wait:/etc/rc.d/rc l1:1:wait:/etc/rc.d/rc l2:2:wait:/etc/rc.d/rc l3:3:wait:/etc/rc.d/rc l4:4:wait:/etc/rc.d/rc l5:5:wait:/etc/rc.d/rc l6:6:wait:/etc/rc.d/rc 0 1 2 3 4 5 6

# Things to run in every runlevel. ud::once:/sbin/update

call the shutdown command to reboot the system when the use does the three fingered salute
ca::ctrlaltdel:/sbin/shutdown -t3 -r now

A powerfail signal will arrive if you have a uninterruptable power supply (UPS) if this happens shut the machine down safely
pf::powerfail:/sbin/shutdown -f -h +2 "Power Failure; System Shutting Down" # If power was restored before the shutdown kicked in, cancel it. pr:12345:powerokwait:/sbin/shutdown -c "Power Restored; Shutdown Cancelled"

Start the login process for the virtual consoles

1:12345:respawn:/sbin/mingetty tty1

David Jones (20.01.00)

Page 324

85321, Systems Administration

Chapter 12: Startup and Shutdown

2:2345:respawn:/sbin/mingetty 3:2345:respawn:/sbin/mingetty 4:2345:respawn:/sbin/mingetty 5:2345:respawn:/sbin/mingetty 6:2345:respawn:/sbin/mingetty

tty2 tty3 tty4 tty5 tty6

If the machine goes into runlevel 5, start X

x:5:respawn:/usr/bin/X11/xdm -nodaemon

The identifier The identifier, the first field, is a unique two character identifier. For inittab entries that correspond to terminals the identifier will be the suffix for the terminals device file. For each terminal on the system a mingetty process must be started by the init process. Each terminal will generally have a device file with a name like /dev/tty??, where the ?? will be replaced by a suffix. It is this suffix that must be the identifier in the /etc/inittab file. Run levels The run levels describe at which run levels the specified action will be performed. The run level field of /etc/inittab can contain multiple entries, e.g. 123, which means the action will be performed at each of those run levels. Actions The actions field describes how the process will be executed. There are a number of pre-defined actions that must be used. Table 10.2 lists and explains them.

David Jones (20.01.00)

Page 325

85321, Systems Administration

Chapter 12: Startup and Shutdown

Action
respawn wait

Purpose restart the process if it finishes init will start the process once and wait until it has finished before going on to the next entry start the process once, when the runlevel is entered perform the process during system boot (will ignore the runlevel field) a combination of boot and wait do nothing specify the default run level execute process during boot and before any boot or bootwait entries executed when init receives the SIGPWR signal which indicates a problem with the power, init will wait until the process is completed execute whenever the ondemand runlevels are called (a b c). When these runlevels are called there is NO change in runlevel. same as powerwait but dont wait (refer to the man page for the action powerokwait) executed when init receives SIGINT signal (usually when someone does CTRL-ALT-DEL
Table 12.2

once boot

bootwait off initdefault sysinit powerwait

ondemand

powerfail

ctrlaltdel

inittab actions

The process The process is simply the name of the command or shell script that should be executed by init. Daemons and Configuration Files init is an example of a daemon. It will only read its configuration file, /etc/inittab, when it starts execution. Any changes you make to /etc/inittab will not influence the execution of init until the next time it starts, i.e. the next time your computer boots. There are ways in which you can tell a daemon to re-read its configuration files. One generic method, which works most of the time, is to send the daemon the HUP signal. For most daemons the first step in doing this is to find out what the process id (PID) is of the daemon. This isnt a problem for init. Why? Its not a problem for init because init always has a PID of 1.

David Jones (20.01.00)

Page 326

85321, Systems Administration

Chapter 12: Startup and Shutdown

The more accepted method for telling init to re-read its configuration file is to use the telinit command. telinit q will tell init to re-read its configuration file. Exercises
12.3.

Add an entry to the /etc/inittab file so that it displays a message HELLO onto your current terminal (HINT: you can find out your current terminal using the tty command). Modify the inittab entry from the previous question so that the message is displayed again and again and.... Take your system into single user mode. Take your system into runlevel 5. What happens? (only do this if you have X Windows configured for your system). Change your system so that it enters this run level when it boots. Reboot your system and see what happens. The wall command is used to display a message onto the terminals of all users. Modify the /etc/inittab file so that whenever someone does the three finger salute (CTRL-ALT-DEL) it displays a message on the consoles of all users and doesnt log out. Examine your inittab file for an entry with the identifier c1. This is the entry for the first console, the screen you are on when you first start your system. Change the entry for c1 so that the action field contains once instead of respawn. Force init to re-read the inittab file and then log in and log out on that console. What happens?

12.4.

12.5. 12.6.

12.7.

12.8.

System Configuration
There are a number of tasks which must be completed once during system startup which must be completed once. These tasks are usually related to configuring your system so that it will operate. Most of these tasks are performed by the /etc/rc.d/rc.sysinit script. It is this script which performs the following operations sets up a search path that will be used by the other scripts obtains network configuration data activates the swap partitions of your system sets the hostname of your system Every UNIX computer has a hostname. You can use the UNIX command hostname to set and also display your machines hostname. sets the machines NIS domain (if you are using one) performs a check on the file systems of your system

David Jones (20.01.00)

Page 327

85321, Systems Administration

Chapter 12: Startup and Shutdown

turns on disk quotas (if being used) sets up plugnplay support deletes old lock and tmp files sets the system clock loads any kernel modules.

Terminal logins
For a user to login there must be a getty process (RedHat Linux uses a program called mingetty, slightly different name but same task) running for the terminal they wish to use. It is one of inits responsibilities to start the getty processes for all terminals that are physically connected to the main machine, and you will find entries in the /etc/inittab file for this. Please note this does not include connections over a network. They are handled with a different method. This method is used for the virtual consoles on your Linux machine and any other dumb terminals you might have connected via serial cables. You should be able see the entries for the virtual consoles in the example /etc/inittab file from above. Exercises
12.9.

When you are in single user mode there is only one way to login to a Linux machine, from the first virtual console. How is this done?

Startup scripts
Most of the services which init starts are started when init executes the system start scripts. The system startup scripts are shell scripts written using the Bourne shell (this is one of the reasons you need to know the bourne shell syntax). You can see where these scripts are executed by looking at the inittab file.
l0:0:wait:/etc/rc.d/rc l1:1:wait:/etc/rc.d/rc l2:2:wait:/etc/rc.d/rc l3:3:wait:/etc/rc.d/rc l4:4:wait:/etc/rc.d/rc l5:5:wait:/etc/rc.d/rc l6:6:wait:/etc/rc.d/rc 0 1 2 3 4 5 6

These scripts start a number of services and also perform a number of configuration checks including checking the integrity of the machines file systems using fsck, mounting the file systems, designating paging and swap areas, checking disk quotas,

David Jones (20.01.00)

Page 328

85321, Systems Administration

Chapter 12: Startup and Shutdown

clearing out temporary files in /tmp and other locations, startin up system daemons for printing, mail, accounting, system logging, networking, cron and syslog.

In the UNIX world there are two styles for startup files: BSD and System V. RedHat Linux uses the System V style and the following section concentrates on this format. Table 12.3 summarises the files and directories which are associated with the RedHat startup scripts. All the files and directories in Table 12.3 are stored in the /etc/rc.d directory. Filename Purpose
rc0.d rc1.d rc2.d rc3.d rc4.d rc5.d rc6.d

directories which contain links to scripts which are executed when a particular runlevel is entered A shell script which is passed the run level. It then executes the scripts in the appropriate directory. Contains the actual scripts which are executed. These scripts take either start or stop as a parameter run once at boot time to perform specific system initialisation steps the last script run, used to do any tasks specific to your local setup that isnt done in the normal SysV setup not always present, used to perform special configuration on any serial ports
Table 12.3 Linux startup scripts

rc
init.d

rc.sysinit

rc.local

rc.serial

The Linux Process

When init first enters a run level it will execute the script /etc/rc.d/rc (as shown in the example /etc/inittab above). This script then proceeds to determine the current and previous run levels kill any services which must be killed start all the services for the new run level.

The /etc/rc.d/rc script knows how to kill and start the services for a particular run level because of the filenames in the directory for each runlevel. The following are the filenames from the /etc/rc.d/rc3.d directory on my system.
[david@beldin rc.d]$ ls rc3.d K20rstatd S05kudzu S20random S40crond S75keytable S99linuxconf K20rusersd S10network S25netfs S45pcmcia S80sendmail S99local K20rwhod S11portmap S30syslog S50inet S85gpm K55routed S16apmd S40atd S60lpd S90xfs

You will notice that all the filenames in this, and all the other rcX.d directories, use the same format.
David Jones (20.01.00) Page 329

85321, Systems Administration

Chapter 12: Startup and Shutdown

[SK]numberService Where number is some integer and Service is the name of a service. All the files with names starting with S are used to start a service. Those starting with K are used to kill a service. From the rc3.d directory above you can see scripts which start services for the Internet (S50inet), PCMCIA cards (S45pcmcia), a and others. The numbers in the filenames are used to indicate the order in which these services should be started and killed. Youll notice that the script to start the Internet services comes before the script to start the Web server; obviously the Web server depends on the Internet services. /etc/rc.d/init.d If we look closer we can see that the files in the rcX.d directories arent really files.
[david@beldin rc.d]$ ls -l rc3.d/S50inet lrwxrwxrwx 1 root root 14 Dec 19 23:57 rc3.d/S50inet -> ../init.d/inet

The files in the rcX.d directories are actually soft links to scripts in the /etc/rc.d/init.d directory. It is these scripts which perform all the work. Starting and stopping The scripts in the /etc/rc.d/init.d directory are not only useful during the system startup process, they can also be useful when you are performing maintenance on your system. You can use these scripts to start and stop services while you are working on them. For example, lets assume you are changing the configuration of your Web server. Once youve finished editing the configuration files you will need to restart the Web server for it to see the changes. One way you could do this would be to follow this example
[root@beldin rc.d]# /etc/rc.d/init.d/httpd stop Shutting down http: [root@beldin rc.d]# /etc/rc.d/init.d/httpd start Starting httpd: httpd

This example also shows you how the scripts are used to start or stop a service. If you examine the code for /etc/rc.d/rc (remember this is the script which runs all the scripts in /etc/rc.d/rcX.d) you will see two lines. One with $i start and the other with $i stop. These are the actual lines which execute the scripts. Lock files All of the scripts which start services during system startup create lock files. These lock files, if they exist, indicate that a particular service is operating. Their main use is to prevent startup files starting a service which is already running.

David Jones (20.01.00)

Page 330

85321, Systems Administration

Chapter 12: Startup and Shutdown

When you stop a service one of the things which has to occur is that the lock file must be deleted. Exercises
12.10.

What would happen if you tried to stop a service when you were logged in as a normal user (i.e. not root)? Try it.

Why wont it boot?

There will be times when you have to reboot your machine in a nasty manner. One rule of thumb used by Systems Administration to solve some problems is "When in doubt, turn the power off, count to ten slowly, and turn the power back on". There will be times when the system wont come back to you, DONT PANIC! Possible reasons why the system wont reboot include hardware problems, Caused by both hardware failure and problems caused by human error (e.g. the power cord isnt plugged in, the drive cable is the wrong way around) defective boot floppies, drives or tapes, damaged file systems, improperly configured kernels, A kernel configured to use SCSI drives wont boot on a system that uses an IDE drive controller. errors in the rc scripts or the /etc/inittab file.

Solutions
The following is a Systems Administration maxim Always keep a separate working method for booting the machine into at least single user mode. This method might be a boot floppy, CD-ROM or tape. The format doesnt matter. What does matter that at anytime you can bring the system up in at least single user mode so you can perform some repairs. A separate mechanism to bring the system up single user mode will enable you to solve most problems involved with damaged file systems, improperly configured kernels and errors in the rc scripts.

Boot and root disks

The concept of boot and root disk are important to understanding how the booting process works and also in creating an alternative boot method for your system. The definitions used are boot disk This is the disk which contains the kernel of your system.

David Jones (20.01.00)

Page 331

85321, Systems Administration

Chapter 12: Startup and Shutdown

root disk The root disk contains the root file system with all the necessary programs and files required for init to start and setup a minimum of services. This includes such things as init, /etc/inittab and associated files, /etc/passwd and other information required to allow people to login plus a whole lot more.

To have a complete alternative boot method you must have both alternative boot and root disks. The alternative boot disk is useful if you have problems with your kernel. The alternative root disk is required when you have problems such as a wrongly configured inittab or a missing /etc/passwd file. It is possible for a single disk to provide both boot and root disk services.

Making a boot and root disk

It is important that you have alternative boot and root disks for your system. There are (at least) two methods you can use to obtain them use the installation disks which come with your distribution of Linux, In order to install Linux you basically have to have a functioning Linux computer. Therefore the installation disk(s) that you used to install Linux provide an alternative boot and root disk. use a rescue disk (set). A number of people have created rescue disks. These are boot and root disk sets which have been configured to provide you with the tools you will need to rescue your system from problems.

The resource materials section for week 7 on the 85321 Web site/CD-ROM contains pointers to two rescue disk sets. Exercises
12.11.

Create a boot and root disk set for your system using the resources on the 85321 Web site/CD-ROM.

Using boot and root

What do you think would happen if you did the following?
rm /etc/inittab

The next time you booted your system you would see something like this on the screen.
INIT: version 2.71 booting INIT: No inittab file found Enter runlevel: 1 INIT: Entering runlevel: 1 INIT: no more processes left in this runlevel

Whats happening here is that init cant find the inittab file and so it cant do anything. To solve this you need to boot the system and replace the missing
David Jones (20.01.00) Page 332

85321, Systems Administration

Chapter 12: Startup and Shutdown

inittab file. This is where the alternative root and boot disk(s) come in handy. To solve this problem you would do the following boot the system with the alternative boot/root disk set login as root perform the following

/> mount t ext2 /dev/hda2 /mnt mount: mount point /mnt does not exist /> mkdir /mnt /> mount t ext2 /dev/hda1 /mnt EXT2-fs warning: mounting unchecked fs, running e2fsck is recommended /> cp /etc/inittab /mnt/etc/inittab /> umount /mnt

A description of the above goes like this Try to mount the usual root file system, the one with the missing inittab file. But it doesnt work. Create the missing /mnt directory. Now mount the usual root file system. Copy the inittab file from the alternative root disk onto the usual root disk. Normally you would have a backup tape which contains a copy of the old inittab file. Unmount the usual root file system and reboot the system.

The aim of this example is to show you how you can use alternative root and boot disks to solve problems which may prevent your system from booting. Exercises
12.12.

Removing the /etc/inittab file from your Linux system will not only cause problems when you reboot the machine. It also causes problems when you try to shut the machine down. What problems? Why? What happens if you forget the root password? Without it you cant perform any management tasks at all. How would you fix this problem? Boot your system in the normal manner and comment out all the entries in your /etc/inittab file that contain the word mingetty. What do you think is going to happen? Reboot your system. Now fix the problem using the installation floppy disks.

12.13.

12.14.

David Jones (20.01.00)

Page 333

85321, Systems Administration

Chapter 12: Startup and Shutdown

Solutions to hardware problems

Some guidelines to solving hardware problems check the power supply and its connections, Dont laugh, there are many cases I know of in which the whole problem was caused by the equipment not being plugged in properly or not at all. check the cables and plugs on the devices, check any fault lights on the hardware, power cycle the equipment (power off, power on), There is an old Systems Administration maxim. If something doesnt work turn it off, count to 10 very slowly and turn it back on again (usually with the fingers crossed). Not only can it solve problems but it is also a good way of relaxing. Of course this is a last resort and in some cases may not be available. For example, if you are in charge of a machine which is required to have 24x7 availability. try rebooting the system without selected pieces of hardware, It may be only one faulty device that is causing the problem. Try isolating the problem device. use any diagnostic programs that are available, or as a last resort call a technician or a vendor.

Damaged file systems

Previous chapters examined file systems and backups. Fixing a damaged file system involved first trying to use the fsck command and if that fails using backups.

Improperly configured kernels

The kernel contains most of the code that allows the software to talk to your hardware. If the code it contains is wrong then your software wont be able to talk to your hardware. In a later chapter on the kernel well explain in more detail why you might want to change the kernel and why it might not work. Suffice to say you must always maintain a working kernel that you can boot your system with.

Shutting down
You should not just simply turn a UNIX computer off or reboot it. Doing so will usually cause some sort of damage to the system especially to the file system. Most of the time the operating system may be able to recover from such a situation (but NOT always). There are a number of tasks that have to be performed for a UNIX system to be shutdown cleanly

David Jones (20.01.00)

Page 334

85321, Systems Administration

Chapter 12: Startup and Shutdown

tell the users the system is going down, Telling them 5 seconds before pulling the plug is not a good way of promoting good feeling amongst your users. Wherever possible the users should know at least a couple of days in advance that the system is going down (there is always one user who never knows about it and complains). signal the currently executing processes that it is time for them to die, UNIX is a multi-tasking operating system. Just because there is no-one logged in this does not mean that there is nothing going on. You must signal all the current running processes that it is time to die gracefully. place the system into single user mode, and perform sync to flush the file systems buffers so that the physical state of the file system matches the logical state.

Most UNIX systems provide commands that perform these steps for you. As computers become more important to the operation of a business systems must have 24x7 availability. Imagine how much money Amazon.com or eBay lose if and when their computers are unavailable. In these situations shutting down a computer usually involves ensuring that there is another computer already running which will take over operations.

Reasons Shutting down

In general, you should try to limit the number of times you turn a computer on or off as doing so involves some wear and tear. It is often better to simply leave the computer on 24 hours a day. In the case of a UNIX system being used for a mission critical application by some business it may have to be up 24 hours a day. Some of the reasons why you may wish to shut a UNIX system down include general housekeeping, Every time you reboot a UNIX computer it will perform some important housekeeping tasks, including deleting files from the temporary directories and performing checks on the machines file systems. Rebooting will also get rid of any zombie processes. general failures, and Occasionally problems will arise for which there is only one resort, shutdown. These problems can include hanging logins, unsuccessful mount requests, dazed devices, runaway processes filling up disk space or CPU time and preventing any useful work being done. system maintenance and additions. There are some operations that only work if the system is rebooted or if the system is in single user mode, for example adding a new device.

Being nice to the users

Knowing of the existence of the appropriate command is the first step in bringing your UNIX computer down. The other step is outlined in the heading for this section. The following command is an example of what not to do.

David Jones (20.01.00)

Page 335

85321, Systems Administration

Chapter 12: Startup and Shutdown

shutdown -h -1 now

Under Linux this results in a message somewhat like this appearing on every users terminal
THE SYSTEM IS BEING SHUT DOWN NOW ! ! ! Log off now or risk your files being damaged.

and the user will almost immediately be logged out. This is not a method inclined to win friends and influence people. The following is a list of guidelines of how and when to perform system shutdowns shutdowns should be scheduled, If users know the system is coming down at specified times they can organise their computer time around those times. perform a regular shutdown once a week, and A guideline, so that the housekeeping tasks discussed above can be performed. If its regular the users get to know when the system will be going down. use /etc/motd. /etc/motd is a text file that contains the message the users see when they first log onto a system. You can use it to inform users of the next scheduled shutdown. /etc/motd is rarely used these days. Its really only visible when you logon to the command line of a UNIX machine. Not to many people do that these days. Alternatives are available including increased use of staff mailing lists.

Commands to shutdown
There are a number of different methods for shutting down and rebooting a system including the shutdown command The most used method for shutting the system down. The command can display messages at preset intervals warning the users that the system is coming down. Most Linux computers are configured so that the three-fingered salute (CTRL-ALT-DEL) will automatically cause the shutdown command to be executed. Refer back to your /etc/inittab file and see if you can see the entry for it. the halt command Logs the shutdown, kills the system processes, executes sync and halts the processor. the reboot command Similar to halt but causes the machine to reboot rather than halting. sending init a TERM signal, init will usually interpret a TERM signal (signal number 15) as a command to go into single user mode. It will kill of user processes and daemons. The

David Jones (20.01.00)

Page 336

85321, Systems Administration

Chapter 12: Startup and Shutdown

command is kill -15 1 (init is always process number 1). It may not work or be safe on all machines. the fasthalt or fastboot commands These commands create a file /fastboot before calling halt or reboot. When the system reboots and the startup scripts find a file /fastboot they will not perform a fsck on the file systems.

The most used method will normally be the shutdown command. It provides users with warnings and is the safest method to use.
shutdown

The format of the command is

shutdown [ -h | -r ] [ -fqs ] [ now | hh:ss | +mins ]

The parameters are

-h

Halt the system and dont reboot.

-r

Reboot the system

-f

Do a fast boot.
-q

Use a default broadcast message.

-s

Reboot into single user mode by creating a /etc/singleboot file.

The time at which a shutdown should occur are specified by the now hh:ss +mins options.

now

Shut down immediately.

hh:ss

Shut down at time hh:ss.

+mins

Shut down mins minutes in the future. The default wait time before shutting down is two minutes.

What happens
The procedure for shutdown is as follows five minutes before shutdown or straight away if shutdown is in less than five minutes The file /etc/nologin is created. This prevents any users (except root) from logging in. A message is also broadcast to all logged in users notifying them of the imminent shutdown.

David Jones (20.01.00)

Page 337

85321, Systems Administration

Chapter 12: Startup and Shutdown

at shutdown time. All users are notified. init is told not to spawn any more getty processes. Shutdown time is written into the file /var/log/wtmp. All other processes are killed. A sync is performed. All file systems are unmounted. Another sync is performed and the system is rebooted.

The other commands

The other related commands including reboot, fastboot, halt, fasthalt all use a similar format to the shutdown command. Refer to the man pages for more information.

Conclusions
Booting and shutting down a UNIX computer is significantly more complex than performing the same tasks with a MS-DOS computer. A UNIX computer should never just be shut off. The UNIX boot process can be summarised into a number of steps the hardware ROM or BIOS performs a number of tasks including loading the bootstrap program, the bootstrap program loads the kernel, the kernel starts operation, configures the system and runs the init process
init

consults the /etc/inittab file and performs a number of necessary actions.

One of the responsibilities of the init process is to execute the startup scripts that, under Linux, reside in the /etc/rc.d directory. It is important that you have at least one other alternative method for booting your UNIX computer. There are a number of methods for shutting down a UNIX computer. The most used is the shutdown command.

Review Questions
12.1 What would happen if the file /etc/inittab did not exist? Find out. 12.2 How would you fix the following problems? The kernel for your Linux computer has been accidentally deleted.

The /etc/fstab file for your system has been moved to /usr/local/etc/fstab.

12.3 Explain each of the following inittab entries

s1:45:respawn:/sbin/agetty 19200 ttyS0 vt100

David Jones (20.01.00)

Page 338

85321, Systems Administration

Chapter 12: Startup and Shutdown

id:5:initdefault:

si:S:sysinit:/etc/rc.d/rc.S

12.4 Your boss has decided that you should prevent people from being able to boot your Linux server via a floppy by modifying the BIOS configuration. Why does this increase the security of the server somewhat? What problems might this approach have?

David Jones (20.01.00)

Page 339

85321, Systems Administration

Chapter 13: Kernel

Chapter
Kernel

The bit of the nut that you eat?

Well, not exactly. The kernel is the core of the operating system; it is the program that controls the basic services that are utilised by user programs; it is this suite of basic services in the form of system calls that make an operating system "UNIX". The kernel is also responsible for: CPU resource scheduling (with the associated duties of process management) Memory management (including the important implementation of protection) Device control (including providing the device-file/device-driver interface) Security (at a device, process and user level) Accounting services (including CPU usage and disk quotas) Inter Process Communication (shared memory, semaphores and message passing)

The Linux Kernel FAQ sums it up nicely with:

The Unix kernel acts as a mediator for your programs. First, it does the memory management for all of the running programs (processes), and makes sure that they all get a fair (or unfair, if you please) share of the processors cycles. In addition, it provides a nice, fairly portable interface for programs to talk to your hardware. Obviously, there is more to the kernels operation than this, but the basic functions above are the most important to know.

Other Resources
Other resources which discuss kernel related matters include HOW-TOs Kernel HOWTO, Kerneld mini-HOWTO, LILO mini-HOWTO, Modules mini HOW-TO The Linux Kernel A book, available from the LDP (on the 85321 CD-ROM), which describes the principles and mechanisms used by the Linux Kernel (version 2.0.33). Linux Kernel Module Programming Guide A book, available from the LDP, which describes how to write kernel modules.

David Jones (20.01.00)

Page 340

85321, Systems Administration

Chapter 13: Kernel

Linux Device Drivers A book from ORelly describing how to write device drivers for Linux. http://www.ora.com/catalog/linuxdrive/ LAME A book from the LDP which includes sections on Linux Kernel Upgrades, Upgrading a Red Hat Stock Kernel, Building a Custom Kernel, and Moving to the Linux 2.2.x Kernels. The RedHat 6.1 Reference Guide Includes a number of sections describing the process for configuring and compiling kernels. The Linux Kernel Archives http://www.kernel.org/ The primary site for the source of the Linux kernel. The International Kernel Patch http://www.kerneli.org/ Where the Linux kernel, with fully-fledged cryptographic support, is distributed (sites in the US cant legally distribute it). Kernel Notes http://www.kernelnotes.org/ A collection of very good links to everything Linux and kernel related.

Why?
Why study the kernel? Isnt that an operating-system-type-thing? What does a Systems Administrator have to do with the internal mechanics of the OS? Lots. UNIX is usually provided with the source for the kernel (there are exceptions to this in the commercial UNIX world). The reason is that this allows Systems Administrators to directly customise the kernel for their particular system. A Systems Administrator might do this because: They have modified the system hardware (adding devices, memory, processors etc.). They wish to optimise the memory usage (called reducing the kernel footprint). The speed and performance of the system may need improvement (eg. modify the quantum per task to suit CPU intensive vs IO intensive systems). This process (along with optimising memory) is called tweaking. Improvements to the kernel can be provided in the form of source code which then allows the Systems Administrator to easily upgrade the system with a kernel recompile. Recompiling the kernel is the process whereby the kernel is reconfigured, the source code is regenerated/recompiled and a linked object is produced. Throughout this chapter the concept of recompiling the kernel will mean both the kernel source code compilation and linkage.
David Jones (20.01.00) Page 341

85321, Systems Administration

Chapter 13: Kernel

How?
In this chapter, we will be going through the step-by-step process of compiling a kernel, a process that includes: Finding out about your current kernel (what version it is and where it is located?) Obtaining the kernel (where do you get the kernel source, how do you unpack it and where do you put it?) Obtaining and reading documentation (where can I find out about my new kernel source?) Configuring your kernel (how is this done, what is this doing?) Compiling your kernel (how do we do this?) Testing the kernel (why do we do this and how?) Installing the kernel (how do we do this?)

But to begin with, we really need to look at exactly what the kernel physically is and how it is generated. To do this, we will examine the Linux kernel, specifically on the x86 architecture.

The lifeless image

The kernel is physically a file that is usually located in the /boot directory. Under Linux, this file is called vmlinuz. On my system, an ls listing of the kernel produced:
bash# ls -al /boot/vml* lrwxrwxrwx 1 root root -rw-r--r-1 root root 14 Jan 2 23:44 /boot/vmlinuz -> vmlinuz-2.0.31 444595 Nov 10 02:59 /boot/vmlinuz-2.0.31

You can see in this instance that the kernel file is actually a link to another file containing the kernel image. The actual kernel size will vary from machine to machine. The reason for this is that the size of the kernel is dependant on what features you have compiled into it, what modifications you've make to the kernel data structures and what (if any) additions you have made to the kernel code. is referred to as the kernel image. At a physical level, this file consists of a small section of machine code followed by a compressed block. At boot time, the program at the start of the kernel is loaded into memory at which point it uncompresses the rest of the kernel.
vmlinuz

This is an ingenious way of making the physical kernel image on disk as small as possible; uncompressed the kernel image could be around one megabyte. So what makes up this kernel?

David Jones (20.01.00)

Page 342

85321, Systems Administration

Chapter 13: Kernel

Kernel gizzards
An umcompressed kernel is really a giant object file; the product of C and assembler linking - the kernel is not an "executable" file (i.e. you just cant type vmlinuz at the prompt to run the kernel). The actual source of the kernel is stored in the /usr/src/linux directory; a typical listing may produce:
[jamiesob@pug jamiesob]$ ls -al /usr/src total 4 drwxr-xr-x 4 root root 1024 drwxr-xr-x 18 root root 1024 lrwxrwxrwx 1 root root 12 drwxr-xr-x 3 root root 1024 drwxr-xr-x 7 root root 1024 Jan Jan Jan Jan Jan 2 2 2 2 2 23:53 23:45 23:44 23:44 23:53 . .. linux -> linux-2.0.31 linux-2.0.31 redhat

/usr/src/linux is a soft link to /usr/src/<whatever linux version> - this means you can store several kernel source

trees - however - you MUST change the soft link of /usr/src/linux to the version of the kernel you will be compiling as there are several components of the kernel source that rely on this. SPECIAL NOTE: If your system doesnt have a /usr/src/linux or a /usr/src/linux* directory (where * is the version of the Linux source) or there is a /usr/src/linux directory but it only has a couple of files then you dont have the source code installed on your machine. The quick solution is to installed the RPM file containing the kernel source code from the Redhat CDROM. You might want to download a more recent version from the Internet. If you are unsure about how to install a RPM file please refer to the Redhat guides. A typical listing of /usr/src/linux produces:
-rw-r--r--rw-r--r--rw-r--r--rw-r--r--rw-r--r--rw-r--r--rw-r--r--rw-r--r--rw-r--r--rw-r--r-drwxr-xr-x drwxr-xr-x drwxr-xr-x drwxr-xr-x drwxr-xr-x drwxr-xr-x drwxr-xr-x drwxr-xr-x drwxr-xr-x drwxr-xr-x drwxr-xr-x -rw-r--r--rwxr-xr-x 1 1 1 1 1 1 1 1 1 1 6 7 13 9 2 2 2 2 2 2 4 1 1 root root root root root root root root root root root root root root root root root root root root root root root root root root root root root root root root root root root root root root root root root root root root root root 2 6282 18458 21861 3221 2869 7042 9817 3114 89712 1024 1024 1024 1024 1024 1024 1024 1024 1024 1024 1024 862 995060 May Aug Dec Aug Dec Jan Aug Aug Aug May May May May May May May May May May Jan May Aug May 12 9 1 17 30 10 17 17 17 12 10 10 12 12 12 12 12 12 12 23 12 17 12 1996 1994 1993 1995 1994 1995 1995 1995 1995 1996 1996 1996 1996 1996 1996 1996 1996 1996 1996 1995 1996 1995 1996 .version CHANGES COPYING CREDITS Configure MAGIC Makefile README README.modules System.map arch/ drivers/ fs/ include/ init/ ipc/ kernel/ lib/ mm/ modules/ net/ versions.mk vmlinux

David Jones (20.01.00)

Page 343

85321, Systems Administration

Chapter 13: Kernel

Take note of the vmlinux (if you have one) file - this is the uncompressed kernel! Notice the size? [vmlinuz is the .z (or compressed) version of vmlinux plus the decompression code] Within this directory hierarchy are in excess of 1300 files and directories. On my system this consists of around 400 C source code files, 370 C header files, 40 Assembler source files and 46 Makefiles. These, when compiled, produce around 300 object files and libraries. At a rough estimate, this consumes around 16 megabytes of space (this figure will vary). While this may seem like quite a bit of code, much of it actually isnt used in the kernel. Quite a large portion of this is driver code; only drivers that are needed on the system are compiled into the kernel, and then only those that are required at run time (the rest can be placed separately in things called modules; we will examine this later). The various directories form logical divisions of the code, especially between the architecture dependant code (linux/arch), drivers (linux/drivers) and architecture independent code. By using grep and find, it is possible to trace the structure of the kernel program, look at the boot process and find out how various parts of it work.

The first incision

An obvious place to start with any large C program is the void main(void) function. If you grep every source file in the Linux source hierarchy for this function name, you will be sadly disappointed. As I pointed out earlier, the kernel is a giant object file - a series of compiled functions. It is NOT executable. The purpose of void main(void) in C is to establish a framework for the linker to insert code that is used by the operating system to load and run the program. This wouldnt be of any use for a kernel it is the operating system! This poses a difficulty - how does an operating system run itself?

Making the heart beat...

In the case of Linux, the following steps are performed to boot the kernel: The boot loader program (e.g. lilo) starts by loading the vmlinuz from disk into memory, then starts the code executing. After the kernel image is decompressed, the actual kernel is started. This part of the code was produced from assembler source; it is totally machine specific. The code for this is located in the /usr/src/linux/arch/i386/kernel/head.S file. Technically at this point the kernel is running. This is the first process (0) and is called swapper. Swapper does some low level checks on the processor, memory and FPU availability, then places the system into protected mode. Paging is enabled.

David Jones (20.01.00)

Page 344

85321, Systems Administration

Chapter 13: Kernel

Interrupts are disabled (every one) though the interrupt table is set up for later use. The entire kernel is realigned in memory (post paging) and some of the basic memory management structures are created. At this point, a function called start_kernel is called. start_kernel is physically located in /usr/src/linux/init/main.c and is really the core kernel function - really the equivalent of the void main(void). main.c itself is virtually the root file for all other source and header files. Tests are run (the FPU bug in Pentium chip is identified amongst other checks including examinations on the DMA chip and bus architecture) and the BogoMip setting is established.
start_kernel

sets up the memory, interrupts and scheduling. In effect, the kernel has now has multi-tasking enabled. The console already has had several messages displayed to it. The kernel command line options are parsed (those passed in by the boot loader) and all embedded device driver modules are initialised. Further memory initialisations occur, socket/networking is started and further bug checks are performed. The final action performed by swapper is the first process creation with whereby the init program is launched. Swapper now enters an infinite idle loop.
fork

It is interesting to note that as a linear program, the kernel has finished running! The timer interrupts are now set so that the scheduler can step in and pre-empt the running process. However, sections of the kernel will be periodically executed by other processes. This is really a huge oversimplification of the kernels structure, but it does give you the general idea of what it is, what it is made up of and how it loads.

Modules
A module is a dynamically loadable object file containing functions for interfacing with a particular device or performing particular tasks. The concept behind modules is simple; to make a kernel smaller (in memory), keep only the bare basics compiled into the kernel. When the kernel needs to use devices, let it load modules into memory. If it doesnt use the modules, let them be unloaded from memory. This concept has also revolutionised the way in which kernels are compiled. No longer do you need to compile every device driver into the kernel; you can simply mark some as modules. This also allows for separate module compilation - if a new device driver is released then it is a simple case of recompiling the module instead of the entire kernel. Modules work by the kernel communicating with a program called kerneld. kerneld is run at boot time just like a normal daemon process. When the kernel notices that a request has come in for the use of a module, it checks if it is loaded in memory. If it is, then the routine is run, however, if not, the kernel gets kerneld to load the module into memory. kerneld also removes the

David Jones (20.01.00)

Page 345

85321, Systems Administration

Chapter 13: Kernel

module from memory if it hasnt been used in a certain period of time (configurable). The concept of modules is a good one, but there are some things you should be aware of: Frequently used devices and devices required in the boot process (like the hard disk) should not be used as modules; these must be compiled into the kernel. While the concept of modules is great for systems with limited memory, should you use them? Memory is cheap - compiling an object into the kernel rather than leaving it as a module may use more memory but is that better than a system that uses its CPU and IO resources to constantly load and unload modules? There are trade offs between smaller kernels and CPU/IO usage with loadable modules. It is probably a good idea to modularise devices like the floppy disk, CDROM and parallel port - these are not used very often, and when they are, only for a short time. It is NOT a good idea to modularise frequently used modules like those which control networking.

There is quite a bit more to kernel modules.

The proc file system

Part of the kernels function is to provide a file-based method of interaction with its internal data structures; it does this via the /proc virtual file system. The /proc file system technically isnt a file system at all; it is in fact a window on the kernels internal memory structures. Whenever you access the /proc file system, you are really accessing kernel memory. So what does it do? Effectively the /proc file system is providing an instant snapshot of the status of the system. This includes memory, CPU resources, network statistics and device information. This data can be used by programs to gather information about a system, an example of which is the top program. top scans through the /proc structures and is able to present the current memory, CPU and swap information, as given below:
7:12pm up 9:40, 1 user, load average: 0.00, 0.00, 0.10 34 processes: 33 sleeping, 1 running, 0 zombie, 0 stopped CPU states: 0.5% user, 0.9% system, 0.0% nice, 98.6% idle Mem: 14940K av, 13736K used, 1204K free, 5172K shrd, 1920K buff Swap: 18140K av, 2304K used, 15836K free PID 789 98 1 84 96 45 6 7 59 61 USER PRI jamiesob 19 root 14 root 1 jamiesob 1 jamiesob 1 root 1 root 1 root 1 root 1 root 1 NI SIZE RES SHRD STAT %CPU %MEM TIME COMMAND 0 102 480 484 R 1.1 3.2 0:01 top 0 1723 2616 660 S 0.3 17.5 32:30 X :0 0 56 56 212 S 0.0 0.3 0:00 init [5] 0 125 316 436 S 0.0 2.1 0:00 -bash 0 81 172 312 S 0.0 1.1 0:00 sh /usr/X11/bin/star 0 45 232 328 S 0.0 1.5 0:00 /usr/sbin/crond -l10 0 27 72 256 S 0.0 0.4 0:00 (update) 0 27 112 284 S 0.0 0.7 0:00 update (bdflush) 0 53 176 272 S 0.0 1.1 0:00 /usr/sbin/syslogd 0 40 144 264 S 0.0 0.9 0:00 /usr/sbin/klogd

David Jones (20.01.00)

Page 346

85321, Systems Administration

Chapter 13: Kernel

63 65 67 73 77

bin root root root root

1 1 1 1 1

0 0 0 0 0

60 58 31 84 107

0 0 0 0 220

188 180 180 208 296

SW SW SW SW S

0.0 0.0 0.0 0.0 0.0

0.0 0.0 0.0 0.0 1.4

0:00 0:00 0:00 0:00 0:00

(rpc.portmap) (inetd) (lpd) (rpc.nfsd) sendmail:accepting

The actual contents of the /proc file system on my system look like:
psyche:~$ ls /proc 1/ 339/ 100/ 45/ 105/ 451/ 108/ 59/ 109/ 6/ 116/ 61/ 117/ 63/ 124/ 65/ 338/ 67/ 7/ 71/ 73/ 77/ 793/ 80/ 84/ 85/ 86/ 87/ 88/ 89/ 90/ 96/ 97/ 98/ cpuinfo devices dma filesystems interrupts ioports kcore kmsg ksyms loadavg meminfo modules net/ pci self/ stat uptime version

Each of the numbered directories store state information of the process by their PID. The self/ directory contains information for the process that is viewing the /proc filesystem, i.e. - YOU. The information stored in this directory looks like:
cmdline cwd - [0303]:132247 environ exe - [0303]:109739 fd/ maps| root - [0303]:2 stat statm (Current command line) (Link to the current working directory) (All environment variables) (Currently executing code) (Directory containing virtual links to file handles) (Memory map structure) (Link to root directory) (Current process statistics) (Current memory statistics)

Most of these files can be cated to the screen. The /proc/filesystems file, when cated, lists the supported file systems. The /proc/cpuinfo file gives information about the hardware of the system:
psyche:~$ cat /proc/cpuinfo cpu : 586 model : Pentium 90/100 mask : E vid : GenuineIntel fdiv_bug : no math : yes hlt : yes wp : yes Integrated NPU : yes Enhanced VM86 : yes IO Breakpoints : yes 4MB Pages : yes TS Counters : yes Pentium MSR : yes Mach. Ch. Exep. : yes CMPXCHGB8B : yes BogoMips : 39.94

Be aware that upgrading the kernel may mean changes to the structure of the /proc file system. This may require software upgrades. Information about this should be provided in the kernel README files.

David Jones (20.01.00)

Page 347

85321, Systems Administration

Chapter 13: Kernel

Exercises
13.1. 13.2.

Find out where kerneld is launched from. What is the purpose of /sbin/lsmod? Try it. What is the contents of the file /proc/modules? Find out where your kernel image is located and how large it is. Examine the /proc file system on you computer. What do you think the /proc/kcore file is? Hint: Have a look at the size of the file.

13.3. 13.4.

Really, why bother?

The most common reason to recompile the kernel is because youve added some hardware and you want the kernel to recognise and (if youre lucky) use it. A very good time to recompile your kernel is after youve installed Linux. The reason for this is that the original Linux kernel provided has extra drivers compiled into it which consume memory. Funnily enough, while the kernel includes a driver for communicating in EBCDIC via a 300 baud modem to a coke machine sitting in the South Hungarian embassy in Cairo [Makefile Question:
Do you want to include support for coke machines located in Cairo? [Y],N,M? Do you want to support South Hungarian Embassy Models [Y],N,M? Support for 300 baud serial link [Y],N,M? Support EBCDIC communication[Y],N,M?

(I might be making this up... :)] ...the kernel, by default, doesnt have support for some very common sound cards and network devices! To be fair, there are good reasons for this (IRQ conflicts etc.) but this does mean a kernel recompile is required. Another good reason to modify the kernel is to customise some of its data structures for your system. Possible modifications include increasing the number of processes the kernel can support (this is a fixed array and cant be set on run time) or modifying the size of certain buffers. One of the great benefits of having the source code for the operating system is that you can play OS-Engineer; it is possible for you to change the scheduling algorithm, memory management scheme or the IPC functionality. While it might be nice to go and do these things, it would be unadvisable to modify the API if you want your programs to still run under Linux. However, there is nothing to stop you adding to the API. You may, for example, wish to add a system call to print "Hello World" to the screen (this would obviously be of great benefit to the rest of the Linux community ;) - this is possible for you to do. Strangely enough, to modify the kernel, you need kernel source code. The actual source can be obtained from a variety of locations. For users who installed Linux from CD ROM, the source can be found within the distribution.

David Jones (20.01.00)

Page 348

85321, Systems Administration

Chapter 13: Kernel

Typically you will actually go back into the installation menu and install only the section that contains the source. However, more often than not, you are actually seeking to upgrade the kernel, so you need the latest kernel source. Because the development of the Linux kernel is an on-going process, new versions of development kernels are constantly being released. It is not unusual for development kernels to be released as often as once per day! The Kernel HOWTO describes some ways to obtain kernels:
You can obtain the source via anonymous ftp from ftp.kernel.org in /pub/linux/kernel/vx.y, where x.y is the version(eg 2.2), and as mentioned before, the ones that end with an odd number are development releases and may be unstable. It is typically labelled linux-x.y.z.tar.gz, where x.y.z is the version number. The sites also typically carry ones with a suffix of .bz2, which have been compressed with bzip2 (these files will be smaller and take less time to transfer). Its best to use ftp.xx.kernel.org where xx is your country code; examples being ftp.at.kernel.org for Austria, and ftp.us.kernel.org for the United States.

Generally you will only want to obtain a "stable" kernel version. Kernels with even minor numbers are the stable kernels. Kernels with odd minor numbers in the version are the development kernels. My current kernel is 2.2.12-20. The minor number in the version number is 2 (the second one) indicating a stable production kernel. The 2.3.X range of kernels is where people are working on new features for the Linux kernel. As of writing this section (January 12, 1999) the latest versions are 2.2.14 for the stable version and 2.3.39 for the development branch. I need to upgrade the kernel on my machine. If you have an extremely new type of hardware then you are often forced into using developmental kernels. There is nothing wrong with using these kernels, but beware that you may encounter system crashes and potential losses of data. During a one year period, the author obtained around twenty developmental kernels, installed them and had very few problems. For critical systems, it is better to stick to known stable kernels. RedHat and other companies which distribute versions of Linux make the latest kernel sources available in distribution specific formats. For example, RedHat primarily uses the RPM package manager. So, youve obtained the kernel source - it will be in one large, compressed file. You know have to unpack the archive. If you are unsure the Linux Kernel HOWTO provides some guidance as does the Redhat guides. A couple of points to note. Some sources install to directories given by the kernel version, not to the linux directory. It may be worth checking on this before you unpack the

David Jones (20.01.00)

Page 349

85321, Systems Administration

Chapter 13: Kernel

source by issuing the following command. It will list all the files and directories that are contained in the source_filename, the kernel archive.
tar -txvf source_filename

This will display a list of files and where they are to be installed. If they are to be installed into a directory other than linux then you must make a symbolic link, called linux in the /usr/src directory to the directory that contains the new source. NEVER just delete your old source - you may need it to recompile your old kernel version if you find the new version isnt working out, though we will discuss other ways round this problem in later sections.

If you are upgrading your kernel regularly, an alternative to constantly obtaining the complete kernel source is to patch your kernel. Patches are basically text files that contain a list of differences between two files. A kernel patch is a file that contains the differences between all files in one version of the kernel to the next. Why would you use them? The only real reason is to reduce download time and space. A compressed kernel source can be extremely large whereas patches are relatively small. Patches are produced as the output from the diff command. For example, given two files:
file1 "vi is a highly exciting program with a wide range of great features I am sure that we will adopt it as part of our PlayPen suite" - Anonymous Multimillionaire Software Farmer file2 "vi is a mildly useless program with a wide range of missing features I am sure that we will write a much better product; we'll call it `Sentence'" - Anonymous Multimillionaire Software Farmer

After executing the command:

diff file1 file2 file2 file3

would contain:

1,2c1,2 < "vi is a highly exciting program with a wide range of great features - I < am sure that we will adopt it as part of our PlayPen suite" --"vi is a mildly useless program with a wide range of missing features - I am sure that we will write a much better product; we'll call it `Sentence'"

To apply a patch, you use the patch command. patch expects a file as a parameter to apply the patch to, with the actual patch file as standard input. Following the previous example, to patch file1 with file3 to obtain file2, wed use the following command:
patch file1 < file3

This command applies the file3 patch to file1. After the command, file1 is the same as file2 and a file called file1.orig has been created as a backup of the original file1.

David Jones (20.01.00)

Page 350

85321, Systems Administration

Chapter 13: Kernel

The Linux HOWTO further explains applying a kernel patch:

Incremental upgrades of the kernel are distributed as patches. For example, if you have version 1.1.45, and you notice that theres a patch46.gz out there for it, it means you can upgrade to version 1.1.46 through application of the patch. You might want to make a backup of the source tree first (make clean and then cd /usr/src; tar zcvf old-tree.tar.gz linux will make a compressed tar archive for you.). So, continuing with the example above, lets suppose that you have patch46.gz in /usr/src. cd to /usr/src and do a zcat patch46.gz | patch -p0 (or patch -p0 < patch46 if the patch isnt compressed). Youll see things whizz by(or flutter by, if your system is that slow) telling you that it is trying to apply hunks, and whether it succeeds or not. Usually, this action goes by too quickly for you to read, and youre not too sure whether it worked or not, so you might want to use the -s flag to patch, which tells patch to only report error messages (you dont get as much of the hey, my computer is actually doing something for a change! feeling, but you may prefer this..). To look for parts which might not have gone smoothly, cd to /usr/src/linux and look for files with a .rej extension. Some versions of patch (older versions which may have been compiled with on an inferior filesystem) leave the rejects with a # extension. You can use find to look for you; find . -name *.rej -print

prints all files who live in the current directory or any subdirectories with a .rej extension to the standard output.

Patches can be obtained from the same sites as the complete kernel sources. A couple of notes about patches: For every new version of the kernel, there is a patch. To upgrade from a kernel version that is five versions behind the version you want, yo have to obtain and apply five patches (e.g. kernel n.n.1 upgrading to n.n.6 requires patches: patch2, patch3, patch4, patch5 and patch6). This gets tedious and is often easier and quicker to simply obtain the entire kernel source again. Patches are forever - when you patch your kernel source, you modify it for good.

Documentation
Every version of the kernel source comes with documentation. There are several "main" files you should read about your current source version including:
/usr/src/linux/README

Instructions on how to compile the kernel

/usr/src/linux/MAINTAINERS

A list of people who maintain the code

/usr/src/linux/Documentation/*

Documentation for parts of the kernel. ALWAYS read the documentation after obtaining the source code for a new kernel, and especially if you are going to be compiling in a new kind of device.

David Jones (20.01.00)

Page 351

85321, Systems Administration

Chapter 13: Kernel

The Linux Kernel-HOWTO is essential reading for anything relating to compiling or modifying the kernel. Linux is the collaborative product of many people. This is something you quickly discover when examining the source code. The code (in general) is neat but sparsely commented; those comments that do exist can be absolutely riotous...well, at least strange :) These are just a selection of the quotes found in the /usr/src/linux/kernel directory:
(fork.c) Fork is rather simple, once you get the hang of it, but the memory management can be a bitch. (exit.c) "I ask you, have you ever known what it is to be an orphan?" (module.c) ... This feature will give you ample opportunities to get to know the taste of your foot when you stuff it into your mouth!!! (schedule.c) The "confuse_gcc" goto is used only to get better assembly code.. Dijkstra probably hates me. To understand this, you have to know who Dijkstra was - remember OS? ... disregard lost ticks for now.. We dont care enough. (sys.c) OK, we have probably got enough memory - let it rip. This needs some heave checking ... I just havent get the stomach for it. I also dont fully understand. Let somebody who does explain it. (time.c) This is ugly, but preferable to the alternatives. ...This is revolting. Bad, bad....

Apart from providing light entertainment, the kernel source comments are an important guide into the (often obscure) workings of the kernel.

Modifying the Kernel

The main reason for recompiling the kernel is to include support for new devices - to do this you simple have to go through the compile process and answer "Yes" to a few questions relating to the hardware you want. However, in some cases you may actually want to modify the way in which the kernel works, or, more likely, one of the data structures the kernel uses. This might sound a bit daunting, but with Linux this is a relatively simple process. For example, the kernel maintains a statically-allocated array for holding a list of structures associated with each process running on the system. When all of
David Jones (20.01.00) Page 352

85321, Systems Administration

Chapter 13: Kernel

these structures are used, the system is unable to start any new processes. This limit is defined within the tasks.h file located in /usr/src/linux/include/linux/ in the form of:
/* * This is the maximum nr of tasks - change it if you need to */ #define NR_TASKS 512 #define MAX_TASKS_PER_USER (NR_TASKS/2) #define MIN_TASKS_LEFT_FOR_ROOT 4

While 512 tasks may seem a lot, on a multiuser system this limit is quickly exhausted. Remember that even without a single user logged on, a Linux system is running between 30 and 50 tasks. For each user login, you can (at peak periods) easily exceed 5 processes per user. Adding this to web server activity (some servers can be running in excess of one hundred processes devoted to processing incoming http requests), mail server, telnet, ftp and other network services, the 512 process limit is quickly reached. Increasing NR_TASKS and recompiling the kernel will allow more processes to be run on the system - the downside to this is that more memory will be allocated to the kernel data area in the form of the increased number of task structures (leaving less memory for user programs). Other areas you may wish to modify include buffer sizes, numbers of virtual terminals and memory structures. Most of these should be modifiable from the .h files found in the kernel source "include" directories. There are, of course, those masochists (like myself) who cant help tinkering with the kernel code and "changing" things (a euphemism for wrecking a nice stable kernel). This isnt a bad thing (there is an entire team of kernel developers world-wide who spend quite a bit of time doing this) but youve got to be aware of the consequences - total system annihilation is one. However, if you feel confident in modifying kernel code, perhaps you should take a quick look at: /usr/src/linux/kernel/sched.c or /usr/src/linux/mm/memory.c (actually, look at the code anyway). These are two of the most important files in the kernel source, the first, sched.c is responsible for task scheduling. The second, memory.c is responsible for memory allocation. Perhaps someone would like to modify memory.c so that when the kernel runs out of memory that the system simply doesnt just "hang" (just one of my personal gripes there... ;)

Compiling the source

As we will discuss in the next section, ALL changes to the kernel should be compiled and tested on DISK before the "new" kernel is installed on the system. The following section will explain how this is done. Obtain the source of the version before the latest kernel. Install the source in the appropriate directory.

David Jones (20.01.00)

Page 353

85321, Systems Administration

Chapter 13: Kernel

Obtain the patch for the latest kernel source and apply it to the source files you previously retrieved.

If you dont have Internet access, do the same thing but using the CDROM. Pick a version of the kernel source, install it, then patch it with the patch for the next version Find out how to generate a patch file based on the differences between more than one file - what is the command that would recursively generate a patch file from two directories? (These puns are getting very sad)

As you are aware (because youve read all the previous chapters and have been paying intense attention), make is a program use to compile source files, generate object files and link them. make actually lets the compilers do the work, however it co-ordinates things and takes care of dependencies. Important tip: Dependencies are conditions that exist due to that fact some actions have to be done after other actions - this is confusing, but wait, it gets worse. Dependencies also relate to the object of the action; in the case of make this relates to if the object (an object can be an object file or a source file) has been modified. For example, using our Humpty scenario: humpty (program) is made up of legs, arms and torso (humpty, being an egg lacked a neck, thus his torso and head are one) - these could be equated to object files. Humptys legs are made up of feet, shins and thighs - again, object files. Humptys feet are made up of toes and other bits (how do you describe an eggs foot???) - these could be equated to source files. To construct humpty, youd start at the simplest bits, like toes, and combine them with other bits to for the feet, then the legs, then finally, humpty. You could not, however, fully assemble the leg without assembling the foot. And if you modified Humptys toes, it doesnt mean youd have to recompile his fingers - youd have to reconstruct the foot object, relink into a new leg object, which youd link with the (pre compiled and unmodified) arms and torso objects - thus forming Humpty.
make, while not specifically designed to handle broken egg reconstruction, does the same thing with source files - based entirely of rules which the user defines within a file called a Makefile. However, make is also clever enough to compile and link only the bits of a program that have been modified since the last compile.

In the case of the kernel, a series of Makefiles are responsible for the kernel construction. Apart from calling compilers and linkers, make can be used for running programs, and in the case of the kernel, one of the programs it calls is an initialisation script. The steps to compile the kernel all make use of the make program. To compile the kernel, you must be in the /usr/src/kernel, and issue (in the following order and as the root user) these commands:
make make make make make config or make menuconfig or make xconfig dep clean zImage or make zdisk zlilo (if the previous was make zImage)

David Jones (20.01.00)

Page 354

85321, Systems Administration

Chapter 13: Kernel

If you are going to be using modules with your kernel, you will require the following two steps:
make modules make modules_install

The following is an explanation of each step.

Configuration
make config is the first phase of kernel recompilation. Essentially make config causes a series of questions to be issued to the user. These questions

relate to what components should be compiled into the kernel. The following is a brief dialog from the first few questions prompted by make config:
psyche:~/usr/src/linux$ make config rm -f include/asm ( cd include ; ln -sf asm-i386 asm) /bin/sh scripts/Configure arch/i386/config.in # # Using defaults found in .config # * * Code maturity level options * Prompt for development and/or incomplete code/drivers (CONFIG_EXPERIMENTAL)[N/y?] n * * Loadable module support * Enable loadable module support (CONFIG_MODULES) [Y/n/?] Y Set version information on all symbols for modules (CONFIG_MODVERSIONS)[N/y/?] Kernel daemon support (e.g. autoload of modules) (CONFIG_KERNELD) [N/y/?] y * * General setup * Kernel math emulation (CONFIG_MATH_EMULATION) [Y/n/?]

A couple of points to note: Each of these questions has an automatic default (capitalised). This default will be changed if you choose another option; i.e. If the default is "N" and you answer "Y" then on the next compile the default will be "Y". This means that you can simply press "enter" through most of the options after your first compile. These first few questions relate to the basic kernel setup: note the questions regarding modules. This is important to answer correctly, as if you wish to include loadable module support, you must do so at this point.

As you progress further through the questions, you will be prompted for choosing support for specific devices, for example:
* * Additional Block Devices * Loopback device support (CONFIG_BLK_DEV_LOOP) [N/y/m/?]

David Jones (20.01.00)

Page 355

85321, Systems Administration

Chapter 13: Kernel

Multiple devices driver support (CONFIG_BLK_DEV_MD) [N/y/?] RAM disk support (CONFIG_BLK_DEV_RAM) [Y/m/n/?] Initial RAM disk (initrd) support (CONFIG_BLK_DEV_INITRD) [N/y/?] XT harddisk support (CONFIG_BLK_DEV_XD) [N/y/m/?]

In this case, note the "m" option? This specifies that the support for a device should be compiled in as a module - in other words, not compiled into the kernel but into separate modules. Be aware that there are quite a few questions to answer in make config. If at any point you break from the program, you must start over again. Some "sections" of make config, like the sound card section, save the results of the first make config in a configuration file; you will be prompted to either reconfigure the sound card options or use the existing configurations file. There are two other methods of configuring the kernel, make menuconfig and make xconfig. The first time you run either of these configuration programs, they will actually be compiled before your very eyes (exciting eh?). menuconfig is just a text based menu where you select the parts of the kernel you want; xconfig is the same thing, just for X-Windows. Using either of these utilities will probably be useful for someone who has never compiled the kernel before, however, for a comprehensive step-by-step selection of kernel components, make config is, in my view, better. You may be wondering what is the result of make config/menuconfig/xconfig? What is actually happening is that small configuration files are being generated to be used in the next step of the process, make dep.

Dependencies
make dep takes the results from make config and "sets up" which parts of the kernel have to be compiled and which dont. Basically this step involves extensive use of sed and awk for string substitution on files. This process may take a few minutes; there is no user interaction at this point.

After running make dep, make clean must be run. Again, this process requires no user interaction. make clean actually goes through the source tree and removes all the old object and temporary files. This process can not be skipped. At this point, we are ready to start the compile process.

Compilation
You have two options at this point; you may either install the kernel on the hard drive of the system and hope it works, or, install the kernel on a floppy disk and test it for a while, then (if it is working) install it on the hard drive. ALWAYS tests your kernel on a floppy disk before installing it as your boot kernel on the hard drive. Why? Simply because if you install your new kernel directly over the one on the hard drive and it doesnt work properly (i.e.. crashes or hangs your system) then you will have difficulty booting your

David Jones (20.01.00)

Page 356

85321, Systems Administration

Chapter 13: Kernel

system (being a well prepared Systems Administrator, youd have a boot disk of course ... ;). To compile your new kernel to disk, you must issue the command:
make zdisk

This will install a bootable kernel on the disk in A:. To boot the system, you simply insert the disk containing the kernel in A:, shut down the system, and let it reboot. The kernel on disk will load into memory, mount your root partition and the system will boot as normal. It is a good idea to run this kernel on disk for at least a few days, if not longer. If something goes wrong and you find your system has become unstable, it is merely a process of removing the disk, rebooting and the system will start up with your old kernel. If you are going to install the kernel directly to the hard disk, then you should issue the commands:
make zImage make zlilo

The first command, make zImage, actually compiles the kernel, the second, make zlilo installs the kernel on whatever root partition you have configured with lilo.

Most systems use lilo as the kernel boot loader. A common misconception is that lilo is only used to boot kernels off hard disks. This is actually incorrect; if lilo is configured (usually done when you installed your system, see "man lilo" for more information on configuring it) to boot the kernel from floppy disk, then running make zlilo will cause a copy of the kernel (and lilo) to be copied onto a disk. However, lilo is usually used to load a kernel form hard disk. The way it works is simple; lilo finds the absolute block/sector address of the kernel image on the disk. It then creates a small program (containing this and other information) and inserts it in the boot sector of the primary hard disk. At boot time, lilo is run, prompting (optionally) the user for the desired operating system to boot. When the choice is made, lilo goes directly to the block/sector of the kernel boot image (or other operating system boot file) and loads it into memory and executes it. The actual compile process (either using make zImage or make zdisk is a lengthy process. A Pentium 100 with 16 megabytes of RAM takes around 15 to 25 minutes to compile the kernel (depending on what has been included). Compiling DEC UNIX on a DEC-Alpha takes around three to four minutes. Have pity for those in the not-so-distant era of the 386 that waited all day for a kernel to recompile. It is quite OK to be recompiling the kernel while other users are logged onto the system; be aware that this will slow the process down and make the system appear VERY slow to the users (unless you have a "really, nice" machine).

David Jones (20.01.00)

Page 357

85321, Systems Administration

Chapter 13: Kernel

If you have decided to use dynamically loadable modules, there are two more commands you must issue:
make modules make modules_install

Note this is done post kernel compile - the useful thing about this is that if you upgrade your modules, you can simply recompile them without the need for a full kernel recompile! After the make zImage/zlilo/zdisk commands and compiling the modules, your kernel is ready to be tested. As previously stated, it is important to test your kernel before using it as your system boot kernel. If you find that the kernel is working normally from disk and it hasnt crashed the system (too much), then you can install the kernel to the hard disk. The easiest way to do this is to go back to the /usr/src/linux directory and type:
make zlilo

This will install the copy of the kernel that was previously compiled to disk (a copy is also kept in the kernel source directory) to the hard drive, or whatever boot device lilo is configured to use.

Common Problems
Did you read the documentation? "If all else fails, read the documentation" this quote is especially true of kernel recompiles. A few common problems that you may be confronted with are:
make can not find the Makefile but it is there!: This is because make is broken. This was a big problem under the 1.2.n kernels when an updated libc.so.x library was released. The problem was that make would not work under 1.3.n kernels that had been recompiled under the 1.2.n versions with the new library; consequently, you couldnt recompile the kernel under the 1.3.n kernels due to the fact make was not working! This has been fixed since, though at the time the solution was to go and get a new version of make. This is a classic example of what can happen when you start upgrading kernels without upgrading all the libraries, compilers and utilities. Always read the README file before recompiling the kernel and make sure you have all the right versions of libraries, compilers and utilities.

dies: This is bad news. It means one of several things: either the config scripts cant find /bin/bash or /bin/sh, some of the source tree is missing, you are not running the program as root or there is something wrong with your system file permissions/links. It is very rare for this to happen with kernels "unpacked straight from the box". If it does happen, check for the previous reasons; if all else fails, go and get another kernel source.
make config/dep/clean make zImage/zdisk

fails: This is one of those sinking feeling moments when you start getting messages during the compile saying "Error: Something didnt

David Jones (20.01.00)

Page 358

85321, Systems Administration

Chapter 13: Kernel

compile/link". Two primary reasons for this are: not running make clean after make dep and not having the correct libraries installed. The kernel compiles and boots but it is unstable: If you are using developmental kernels, this comes with the territory: because developmental kernels can be unstable. If, however, you are using a known "stable" kernel, then the reason is most likely a hardware conflict. Typical culprits are sound cards and network cards. Remove these from the kernel and recompile. You should then examine the documentation on the offending devices to see what the conflict is. Other reasons for kernel instability include compiling in support for devices you dont have (this is rare but can happen) or the fact that youve just discovered a "real" bug in the kernel - in which case the README documentation will assist you in locating the right person to talk to.

If you are still encountering problems, you should examine the newsgroup archives concerned with Linux. There are also several useful mailing lists and web sites that can assist you with kernel problems. Exercises
13.5.

Modify the kernel so that the maximum number of tasks it can run is 50. Compile this kernel to a floppy disk. See how long it takes to use all these processes up. Modify your kernel so that the kernel version message (seen on boot time) contains your name. Hint: /usr/src/linux/init contains a file called version.c - modify a data structure in this. Recompile your own kernel, including only the components you need. For those components that you need but dont use very oftem, compile them in as modules. Initially boot the kernel from disk, then install it on your hard disk.

13.6.

13.7.

Conclusions
In this chapter we have examined: What is a kernel? Why would a Systems Administrator recompile a kernel? What makes up a modern kernel? How would you obtain a kernel? Why and how would you modify the kernel source? How is a kernel configured and recompiled? Why should a kernel be tested? How is a kernel installed? Issues associated with the modern Linux kernel

David Jones (20.01.00)

Page 359

85321, Systems Administration

Chapter 13: Kernel

Further information of the Linux kernel can be obtained from the Linux Kernel HOWTO and the other resources mentioned at the beginning of this chapter.

Review Questions
13.1. Describe the functions of the kernel; explain the difference between a kernel that uses modules and one that doesnt. 13.2. You have added a D-Link ethernet card to your laptop (a D-Link ethernet card runs via the parallel port). Describe the steps youd perform to allow the system to recognise it. Would you compile support for this module directly into the kernel or make it a module? Why/Why not? 13.3. You wish to upgrade the kernel on an older system (ver 1.2.n) to the latest kernel. What issues should you consider? What problems could occur with such an upgrade; how would you deal with these?

David Jones (20.01.00)

Page 360

85321, Systems Administration

Chapter 14: Automation and Observation

Chapter
Automation and Observation

Introduction
Setting up a machine is one part of Systems Administration. Another somewhat more important part is keeping that machine going. Central to achieving this aim are the two activities we look at in this chapter (there are more) 1. Automation Any tasks which occurs more than once must be automated. The primary tool on UNIX systems for achieving this are shell programs. This chapter looks at the use of cron (the Linux scheduler) for automatically scheduling shell programs and other tasks. 2. Observation People will do things to your computer. Some of them will do nasty things. Observation is the act of keeping an eye on your machine so you know there is something wrong with it before the users complain about something not working. We look at observation from two perspectives: historical and current. Historical observation tells you what has happened on your system. Current observation tells you what is happening now.

Other Resources
Other resources which discuss similar topics include LAME A section called Automatic tasks with Cron and Crontab files. USAIL A section on automating tasks with Cron. http://www.uwsg.indiana.edu/usail/index/automate.html

Automation and cron

A number of the responsibilities of a System Administrator are automated tasks that must be carried out at the regular times every day, week or hour. Examples include, early every morning freeing up disk space by deleting entries in the /tmp directory, performing backups every night or compressing and archiving log files. Most of these responsibilities require no human interaction other than to start the command. Rather than have the Administrator start these jobs manually,

David Jones (20.01.00)

Page

85321, Systems Administration

Chapter 14: Automation and Observation

UNIX provides a mechanism that will automatically carry out certain tasks at set times. This mechanism relies on the cron system. For example, the mirror of the Linux Documentation Project (LDP) on the 85321 website is kept up to date with a cron job (a task scheduled with cron). This particular cron job, run every sunday night, connects to the central LDP site and transfers any updated data.

Components of cron
The cron system consists of the following three components crontab (the cron configuration) files These are the files which tell the cron system which tasks to perform and when. the crontab command This is the command used to modify the crontab files. Even though the crontab files are text files they should not be edited using a text editor. the daemon, crond The cron daemon is responsible for reading the crontab file and then performing the required tasks at the specified times. The cron daemon is started by a system startup file.

crontab crontab

format

files are text files with each line consisting of 6 fields separated by spaces. The first five fields specify when to carry out the command and the sixth field specifies the command. Table 14.1, on the following page, outlines the purpose of each of the fields. Field minute minute of the hour, 00 to 59 hour day month hour of the day, 00 to 32 (military time) day of the month, 1 to 31 month of the year, 1 to 12 Purpose

weekday day of the week, Linux uses three letter abbreviations, sun, mon, tue,.... command The actual command to execute
Table 14.1

crontab fields

Comments can be used and are indicated using the # symbol just as with shell programs. Anything that appears after a # symbol until the end of that line is considered a comment and is ignored by crond. The five time fields can also use any one of the following formats an asterix that matches all possible values, a single integer that matches that exact value,

David Jones (20.01.00)

Page

85321, Systems Administration

Chapter 14: Automation and Observation

a list of integers separated by commas (no spaces) used to match any one of the values two integers separated by a dash (a range) used to match any value within the range.

For example Some example crontab entries include (all but the first two examples are taken from the Linux man page for crontab)
0 * * * * echo Cuckoo Cuckoo > /dev/console 2>&1

Every hour (when minutes=0) display Cuckoo Cuckoo on the system console.
30 9-17 * 1 sun,wed,sat echo date >> /date.file 2>&1

At half past the hour, between 9 and 5, for every day of January which is a Sunday, Wednesday or Saturday, append the date to the file date.file
0 */2 * * * date

Every two hours at the top of the hour run the date command
0 23-7/2,8 * * * date

Every two hours from 11p.m. to 7a.m., and at 8a.m.

0 11 4 * mon-wed date

At 11:00 a.m. on the 4th and on every mon, tue, wed

0 4 1 jan * date

4:00 a.m. on january 1st

0 4 1 jan * date >> /var/log/messages 2>&1

Once an hour, all output appended to log file Output When commands are executed by the crond daemon there is no terminal associated with the process. This means that standard output and standard error, which are usually set the terminal, must be redirected somewhere else. In this case the output is emailed to the person whos crontab file the command appears. It is possible to use I/O redirection to redirect the output of the commands to files. Some of the examples above use output redirection to send the output of the commands to a log file. Exercises
14.1.

Write crontab entries for the following. - run the program date every minute of every day and send the output to a file called date.log - remove all the contents of the directory /tmp at 5:00am every morning - execute a shell script /root/weekly.job every Wednesday - run the program /root/summary at 3, 6 and 9 pm for the first five days of a month

David Jones (20.01.00)

Page

85321, Systems Administration

Chapter 14: Automation and Observation

Creating crontab files

files should not be modified using an editor instead they should be created and modified using the crontab command. Refer for the manual page for crontab for more information but the following are two of the basic methods for using the command.
crontab

1. crontab [file] 2. crontab [-e | -r | -l ] [username] Version 1 is used to replace an existing crontab file with the contents of standard input or the specified file. Version 2 makes use of one of the following command line options
-e

Allows the user to edit the crontab file using an editor (the command will perform some additional actions to make it safe to do so)
-r

Remove the users crontab file

-l

Display the users crontab file onto standard output By default all actions are carried out on the users own crontab file. Only the root user can specify another username and modify that users crontab file. Exercise
14.2.

Using the crontab command to add the following to your crontab file and observe what happens. run the program date every minute of every day and send the output to a file called date.log

Current Observation
A part of the day to day operation of a system is keeping an eye on the systems current state. This section introduces a number of commands and tools that can be used to examine the current state of the system. The tools are divided into two sections based on what they observe. The sections are disk and file system observation, and The commands du and df process observation and manipulation. The commands ps, kill, nice and top.

David Jones (20.01.00)

Page

85321, Systems Administration

Chapter 14: Automation and Observation

df df summarises that amount of free disk space. By default df will display the following information for all mounted file systems

total number of disk blocks, number of disk blocks used, number available percentage of disk blocks used, and where the file system is mounted.

df also has an option, -i to display Inode usage rather than disk block usage. What an Inode is will be explained in a later chapter. Simply every file that is created must have an Inode. If all the Inodes are used you cant create anymore files. Even if you have disk space available.

The -T option will cause df to display each file systems type. Exercise
14.3.

Use the df command to answer the following questions - how many partitions do you have mounted - how much disk space do you have left on your Linux partition - how many more files can you create on your Linux partition

du

The du command is used to discover the amount of disk space used by file or directory. By default du reports file size as a number of 1 kilobyte blocks. There are options to modify the command so it reports size in bytes (-b) or kilobytes (-k). If you use du on a directory it will report back the size of each file and directory within it and recursively descend down any sub-directories. The -s switch is used to produce the total amount of disk used by the contents of a directory. There are other options that allow you to modify the operation of du with respect to partitions and links. Exercise
14.4.

Use the du command to answer the following questions - how many blocks does the /etc/passwd file use, - how large (in bytes) is the /etc/passwd file, - how disk space is used by the /etc/ directory, the usr directory

David Jones (20.01.00)

Page

85321, Systems Administration

Chapter 14: Automation and Observation

System Status
Table 14.2 summarises some of the commands that can be used to examine the current state of your machine. Some of the information they display includes amount of free and used memory, the amount of time the system has been up, the load average of the system, Load average is the number processes ready to be run and is used to give some idea of how busy your system is. the number of processes and amount of resources they are consuming.

Some of the commands are explained below. For those that arent use your systems manual pages to discover more. Command
free uptime

Purpose display the amount of free and used memory how long has the system been running and what is the current load average one off snap shot of the current processes continual listing of current processes display system information including the hostname, operating system and version and current date and time The Gnome system monitor, a GUI which provides a view of running processes, memory and file system usage (see chapter 5)
Table 14.2 System status commands

Ps/pstree top uname

gtop

ps

The ps command displays a list of information about the process that were running at the time the ps command was executed.
ps has a number of options that modify what information it displays. Table 14.3 lists some of the more useful or interesting options that the Linux version of PS supports.

Table 14.4 explains the headings used by ps for the columns it produces. For more information on the ps command you should refer to the manual page.

David Jones (20.01.00)

Page

85321, Systems Administration

Chapter 14: Automation and Observation

Option
l u m a

Purpose long format displays username (rather than uid) and the start time of the process display process memory info display processes owned by other users (by default ps only shows your processes) shows processes that arent controlled by a terminal use a tree format to show parent/child relationships between processes dont truncate lines to fit on screen
Table 14.3

x f w

ps options

Field
NI SIZE RSS STAT

Purpose the nice value memory size of the processes code, data and stack kilobytes of the program in memory (the resident set size) the status of the process (R-runnable, S-sleeping, D-uninterruptable sleep, T-stopped, Z-zombie) the controlling terminal
Table 14.4 ps fields

TTY

Exercise
14.5.

Use the ps command to answer the following questions - how many processes do you currently own - how many processes are running on your system - how much RAM does the ps command use - whats the current running process

top
provides a one-off snap shot of the processes on your system. For an ongoing look at the processes Linux generally comes with the top command. It also displays a collection of other information about the state of your system including
ps

uptime, the amount of time the system has been up the load average, the total number of processes, percentage of CPU time in user and system mode,
Page

David Jones (20.01.00)

85321, Systems Administration

Chapter 14: Automation and Observation

memory usage statistics statistics on swap memory usage

Refer to the man page for top for more information.

top is not a standard UNIX command however it is generally portable and available for most platforms.

displays the process on your system ranked in order from the most CPU intensive down and updates that display at regular intervals. It also provides an interface by which you can manipulate the nice value and send processes signals.
top

The nice value The nice value specifies how "nice" your process is being to the other users of the system. It provides the system with some indication of how important the process is. The lower the nice value the higher the priority. Under Linux the nice value ranges from -20 to 19. By default a new process inherits the nice value of its parent. The owner of the process can increase the nice value but cannot lower it (give it a higher priority). The root account has complete freedom in setting the nice value.

nice
The nice command is used to set the nice value of a process when it first starts.

renice
The renice command is used to change the nice value of a process once it has started. Signals When you hit the CTRL-C combination to stop the execution of a process a signal (the TERM signal) is sent to the process. By default many processes will terminate when they receive this signal The UNIX operating system generates a number of different signals. Each signal has an associated unique identifying number and a symbolic name. Table 14.6 lists some of the more useful signals used by the Linux operating system. There are 32 in total and they are listed in the file
/usr/include/linux/signal.h

Chapter 5 has some additional discussion about signals.

SIGHUP

The SIGHUP signal is often used when reconfiguring a daemon. Most daemons will only read the configuration file when they startup. If you modify the configuration file for the daemon you have to force it to re-read the file. One method is to send the daemon the SIGHUP signal.
David Jones (20.01.00) Page

85321, Systems Administration

Chapter 14: Automation and Observation

SIGKILL

This is the big "dont argue" signal. Almost all processes when receiving this signal will terminate. It is possible for some processes to ignore this signal but only after getting themselves into serious problems. The only way to get rid of these processes is to reboot the system. Symbolic Name Numeric identifier
SIGHUP SIGKILL SIGTERM 1 9 15

Purpose hangup the kill signal software termination

Table 14.5 Linux signals

kill
The kill command is used to send signals to processes. The format of the kill command is
kill [-signal] pid

This will send the signal specified by the number signal to the process identified with process identifier pid. The kill command will handle a list of process identifiers and signals specified using either their symbolic or numeric formats. By default kill sends signal number 15 (the TERM signal).

Historical Observation
There will be times when you want to reconstruct what happened in the lead up to a problem. Situations where this might be desirable include you believe someone has broken into your system, one of the users performed an illegal action while online, and the machine crashed mysteriously at some odd time. You want to track how much a particular system or resource is used, e.g. a Web server. This can also be useful in justifying to management the need for additional resources. logging, and The recording of certain events, errors, emergencies. accounting. Recording who did what and when.

This is where

become useful.

David Jones (20.01.00)

Page

85321, Systems Administration

Chapter 14: Automation and Observation

This section examines the methods under Linux by which logging and accounting are performed. In particular it will examine the syslog system, process accounting, and login accounting.

Managing log and accounting files

Both logging and accounting tend to generate a great deal of information especially on a busy system. One of the decisions the Systems Administrator must make is what to do with these files. Options include dont create them in the first place, The head in the sand approach. Not a good idea. keep them for a few days, then delete them, and If a problem hasnt been identified within a few days then assume there is no reasons to keep the log files. Therefore delete the existing ones and start from scratch. keep them for a set time and then archive them. Archiving these files might include compressing them and storing them online or copying them to tape.

logrotate Linux systems come with a command called logrotate. As the name suggests this command is used to aid in the management of log files. logrotate allows the automatic rotation, compression, removal and mailing of log files on a daily, weekly, monthly or size basis. On Redhat Linux the logrotate command is configured with the file /etc/logrotate.conf.

Centralise
If you are managing multiple computers it is advisable to centralise the logging and accounting files so that they all appear on the one machine. This makes maintaining and observing the files easier. The syslog system (discussed below) provides this ability.

Security
Since log files are your record of what has occured it is important that they are stored securely. This is another reason for keeping the log files for computers on a single, very secure system. One of the first things someone breaking into your system will attempt to do is to modify the log files so that their actions dont appear. Keeping log files safe is especially important as in some situations they may be required as legal evidence.

David Jones (20.01.00)

Page

85321, Systems Administration

Chapter 14: Automation and Observation

Look at them
Late in 1999 the disk drives in the computer which acts as the Web server for a certain faculty at a certain University failed. It appears the RAID controller for the disk had detected and started logging errors with the disk about five months earlier. The problem was that no-one was reading the log file. It is important that log files actually be read.

Logging
The ability to log error messages or the actions carried out by a program or script is fairly standard. On earlier versions of UNIX each individual program would have its own configuration file that controlled where and what to log. This led to multiple configuration and log files that made it difficult for the Systems Administrator to control and each program had to know how to log.
syslog

The syslog system was devised to provide a central logging facility that could be used by all programs. This was useful because Systems Administrators could control where and what should be logged by modifying a single configuration file and because it provided a standard mechanism by which programs could log information. Components of syslog The syslog system can be divided into a number of components default log file, On many systems messages are logged by default into the file
/var/log/messages

the syslog message format, the application programmers interface, The API programs use to log information. the daemon, and The program that directs logging information to the correct location based on the configuration file. the configuration file. Controls what information is logged and where it is logged.

Exercise
14.6.

Examine the contents of the file /var/log/messages. You will probably have to be the root user to do so. One useful piece of information you should find in that file is a copy of the text that appears as Linux boots.

David Jones (20.01.00)

Page

85321, Systems Administration

Chapter 14: Automation and Observation

syslog syslog

message format

uses a standard message format for all information that is logged. This format includes a facility, The facility is used to describe the part of the system that is generating the message. Table 14.3 lists some of the common facilities. a level, The level indicates the severity of the message. In lowest to highest order the levels are debug info notice warning err crit alert emerg and a string of characters containing a message. Facility Source
kern mail lpr daemon auth

the kernel the mail system the print system a variety of system daemons the login authentication system
Table Common 14.6

syslog facilities

syslogs

API

In order for syslog to be useful application programs must be able to pass messages to the syslog daemon so it can log the messages according to the configuration file.. There are at least two methods which application programs can use to send messages to syslog. These are:
logger,

logger is a UNIX command. It is designed to be used by shell programs which wish to use the syslog facility. the syslog API. The API (application program interface) consists of a set of the functions (openlog syslog closelog) which are used by programs written in compiled languages such as C and C++. This API is defined in the syslog.h file. You will find this file in the system include directory /usr/include.

Exercises
14.7.

Examine the manual page for logger. Use logger from the command line to send a message to syslog Examine the manual page for openlog and write a C program to send a message to syslog

14.8.

David Jones (20.01.00)

Page

85321, Systems Administration

Chapter 14: Automation and Observation

syslogd

is the syslog daemon. It is started when the system boots by one of the startup scripts. syslogd reads its configuration file when it startups or when it receives the HUP signal. The standard configuration file is /etc/syslog.conf.
syslogd syslogd receives logging messages and carries out actions as specified in the configuration file. Standard actions include

appending the message to a specific file, forwarding the message to the syslogd on a different machine, or display the message on the consoles of all or some of the logged in users.

/etc/syslog.conf

By default syslogd uses the file /etc/syslog.conf as its configuration file. It is possible using a command line parameter of syslogd to use another configuration file. A syslog configuration file is a text file. Each line is divided into two fields separated by one or more spaces or tab characters a selector, and Used to match log messages. an action. Specifies what to do with a message if it is matched by the selector

The selector The selector format is facility.level where facility and level level match those terms introduced in the syslog message format section from above. A selector field can include multiple selectors separated by ; characters multiple facilities, separated by a , character, for a single level an * character to match all facilities or levels

The level can be specified with or without a =. If the = is used only messages at exactly that level will be matched. Without the = all messages at or above the specified level will be matched.
syslog.conf

actions

The actions in the syslog configuration file can take one of four formats a pathname starting with / Messages are appended onto the end of the file. a hostname starting with a @ Messages are forwarded to the syslogd on that machine.
Page

David Jones (20.01.00)

85321, Systems Administration

Chapter 14: Automation and Observation

a list of users separated by commas Messages appear on the screens of those users if they are logged in. an asterix Messages are displayed on the screens of all logged in users.

For example The following is an example syslog configuration file taken from the Linux manual page for syslog.conf
# Log all kernel messages to the console. # Logging much else clutters up the screen. #kern.* /dev/console # Log anything (except mail) of level info or higher. # Dont log private authentication messages! *.info;mail.none;authpriv.none /var/log/messages # The authpriv file has restricted access. authpriv.* /var/log/secure # Log all the mail messages in one place. mail.* /var/log/maillog # Everybody gets emergency messages, plus log them on another # machine. *.emerg * # Save mail and news errors of level err and higher in a # special file. uucp,news.crit /var/log/spooler

Exercise
14.9.

A common problem on many systems are users who consume too much disk space. One method to deal with this is to have a script which regularly checks on disk usage by users and reports those users who are consuming too much. The following is one example of a script to do this.
#!/bin/bash # global constant # DISKHOGFILE holds the location of the file defining each users # maximum disk space DISKHOGFILE="disk.hog" # OFFENDERFILE specifiesl where to write information about offending # users OFFENDERFILE="offender" space_used() # accept a username as 1st parameter # return amount of disk space used by the users home directory # in a variable usage { # home directory is the sixth field in /etc/passwd the_home=grep ^$1: /etc/passwd | cut -d: -f6 # du uses a tab character to seperate out its fields # were only interested in the first one

David Jones (20.01.00)

Page

85321, Systems Administration

Chapter 14: Automation and Observation

usage=du -s $the_home | cut -f1 } # # Main Program # while read username max_space do space_used $username if [ $usage -gt $max_space ] then echo $username has a limit of $max_space and has used $used $OFFENDERFILE fi done < $DISKHOGFILE

Modify this script so that it uses the syslog system rather than displaying its output onto standard output.
14.10.

Configure syslog so the messages from the script in the previous question are appended to the logfile /var/log/disk.hog.messages and also to the main system console.

Accounting
Accounting was developed when computers were expensive resources and people were charged per command or CPU time. In todays era of cheap, powerful computers its rarely used for these purposes. One thing accounting is used for is as a source of records about the use of the system. Particular useful if someone is trying, or has, broken into your system. In this section we will examine login accounting. Keeping a track of who has logged into the system and how long they were logged in. process accounting

Login accounting
The file /var/log/wtmp is used to store the username, terminal port, login and logout times of every connection to a Linux machine. Every time you login or logout the wtmp file is updated. This task is performed by init.
last

The last command is used to view the contents of the wtmp file. There are options to limit interest to a particular user or terminal port. Exercise
14.11.

Use the last command to - count how many logins there have been since the current wtmp file was created, - how many times has the root user logged in
Page

David Jones (20.01.00)

85321, Systems Administration

Chapter 14: Automation and Observation

ac

The last command provides rather rudimentary summary of the information in the wtmp file. As a Systems Administrator it is possible that you may require more detailed summaries of this information. For example, you may desire to know the total number of hours each user has been logged in, how long per day and various other information. The command that provides this information is the ac command. Installing ac It is possible (even likely) that you will not have the ac command installed. The ac command is part of the psacct package. You may have to install it. Refer to the RedHat guides for information on how to do this..
14.12.

Use the ac command to - find the total number of hours you were logged in as the root user - find the average number of hours per login for all users - find the daily totals for root

Process accounting
Also known as CPU accounting, process accounting records the elapsed CPU time, average memory use, I/O summary, the name of the user who ran the process, the command name and the time each process finished. You may also need to install process accounting. Turning process accounting on Process accounting does not occur until it is turned on using the accton command.
accton /var/log/acct

Where /var/log/acct is the file in which the process accounting information will be stored. The file must already exist before it will work. You can use any filename you wish but many of the accounting utilities rely on you using this file.
lastcomm lastcomm is used to display the list of commands executed either for everyone, for particular users, from particular terminals or just information about a particular command. Refer to the lastcomm manual page for more information.
[root@beldin /proc]# lastcomm david netscape david tty1 [root@beldin /proc]# lastcomm ttyp2 lastcomm root ttyp2 ls root ttyp2 ls root ttyp2 accton root ttyp2 0.02 secs Sun Jan 25 16:26 0.55 0.03 0.02 0.01 secs secs secs secs Sun Sun Sun Sun Jan Jan Jan Jan 25 25 25 25 16:21 16:21 16:21 16:21

David Jones (20.01.00)

Page

85321, Systems Administration

Chapter 14: Automation and Observation

The sa command The sa command is used to provide more detailed summaries of the information stored by process accounting and also to summarise the information into other files.
[root@beldin /proc]# /usr/sbin/sa -a 66 0.19re 0.25cp 6 0.01re 0.16cp 8 0.00re 0.04cp 17 0.00re 0.01cp 6 0.01re 0.01cp 1 0.00re 0.01cp 5 0.01re 0.01cp 1 0.15re 0.01cp 6 0.01re 0.01cp 5 0.00re 0.00cp 1 0.00re 0.00cp 2 0.00re 0.00cp 1 0.00re 0.00cp 1 0.00re 0.00cp 1 0.00re 0.00cp 1 0.00re 0.00cp 1 0.00re 0.00cp 1 0.00re 0.00cp 2 0.00re 0.00cp

cat lastcomm ls man troff less in.ftpd sh gunzip grotty sa groff gtbl gzip sh* netscape* accton bash*

Refer to the manual pages for the sa command for more information.

So what?
This section has given a very brief overview of process and login accounting and the associated commands and files. What use do these systems fulfil for a Systems Administrator? The main one is that they allow you to track what is occurring on your system and who is doing it. This can be useful for a number of reasons tracking which users are abusing the system figuring out what is normal for a user If you know that most of your users never use commands like sendmail and the C compilers (via process accounting) and then all of a sudden they start using this might be an indication of a break in. justifying to management the need for a larger system Generally management wont buy you a bigger computer just because you want one. In most situations you will have to put together a case to justify why the additional expenditure is necessary. Process and login account could provide some of the necessary information.

David Jones (20.01.00)

Page

85321, Systems Administration

Chapter 14: Automation and Observation

Conclusions
The cron system is used to automatically perform tasks at set times. Components of the cron system include the daemon, crond, Which actually performs the specified tasks. files, and That specify the when and what.
crontab

the crontab command. Used to manipulate the crontab files.

Useful commands for examining the current status of your systems file system include df and du. Commands for examining and manipulating processes include ps, kill, renice, nice and top. Other "status" commands include free, uptime and uname. is a centralised system for logging information about system events. Its components include
syslog

an API and a program (logger) by which information can be logged, the syslogd daemon that actually performs the logging, and the /etc/syslog.conf that specifies what and where logging information should be logged.

Login accounting is used to track when, where and for how long users connect to your system. Process accounting is used to track when and what commands were executed. By default Linux does not provide full support for either form of accounting (it does offer some standard login accounting but not the extra command sac). However there are freely available software distributions that provide Linux this functionality. Login accounting is performed in the /var/log/wtmp file that is used to store the details of every login and logout from the system. The last command can be used to view the contents of the binary /var/log/wtmp file. The nonstandard command sac can be used to summarise this information into a number of useful formats. Process accounting must be turned on using the accton command and the results can be viewed using the lastcomm command. Both logging and accounting can produce files that grow to some considerable size in a short amount of time. The Systems Adminstrator must implement strategies to deal with these log files. Either by ignoring and deleting them or by saving them to tape.

David Jones (20.01.00)

Page

85321, Systems Administration

Chapter 14: Automation and Observation

Review Questions
14.1 Explain the relationship between each of the following
crond, crontab

files and the crontab command, and /etc/syslog.conf and sac

syslogd, logger

/var/adm/wtmp, last

14.2 You have just modified the /etc/syslog.conf file. Will your changes take effect immediately? If not what command would you use to make the modifications take effect? How could you check that the modifications are working? 14.3 Write crontab entries to achieve the following run the script /usr/local/adm/bin/archiveIt every Monday at 6 am run a script /usr/local/adm/bin/diskhog on Monday, Wednesday and Friday at 6am, 12pm, 4pm

14.4 You are a script kiddie (a deogatory name for someone who breaks into computers using very simplistic and automated approaches) who has just broken into my Linux computer. You need to answer the following How do you find what form of logging and accounting I have installed? What can you do to cover the fact that you have broken into my system?

David Jones (20.01.00)

Page

85321, Systems Administration

Chapter 15, Networks: The Connection

Chapter
Networks: The Connection

Introduction
Connecting computers to networks and managing those networks are probably the most important, or at least the most hyped, areas of computing at the moment. This and the following chapter introduce the general concepts associated with TCP/IP-based networks and in particular the knowledge required to connect and use Linux computers to those networks. This chapter examines how you connect a Linux machine and configure it to provide basic network connections and services for other machines. Higher level network applications, such as file sharing and Web servers, and how they work and what you can do with them is the topic for the following chapter. This chapter Overview Provides an overview of connecting a Linux machine to a network. TCP/IP Basics A brief introduction to the fundamentals of TCP/IP networking. Hardare Quick coverage of the hardware which can be used to networking Kernel support Network configuration

Other Resources
As you might expect there is a large amount of information about creating and maintaining TCP/IP networks on the Internet. The following is a small list of some of that material HOWTOs Linux Networking-HOWTO which describes how to install and configure the Linux networking software and associated tools. Linux Networking Overview HOWTO provides an overview of the networking capabilities of Linux and provides pointers to further information. Multicast over TCP/IP HOWTO, DNS HOWTO covers the configuration of the Domain Name Service on Linux, Ethernet HOWTO, IPX HOWTO covers the installation on Linux of the network protocol used by Novell, IP Masquerade HOWTO, ISP Hookup HOWTO, PLIP Install HOWTO covers how to connect Linux boxes using null parallel cables, PPP HOWTO, Asymmetric Digital Subscriber Loop mini-HOWTO, Bridge mini-HOWTO,
Page 380

David Jones (20.01.00)

85321, Systems Administration

Chapter 15, Networks: The Connection

Bridge+Firewall mini-HOWTO, Cipe+Masquerading mini-HOWTO, IP Alias mini-HOWTO, IP Subnetworking mini-HOWTO, Leased Line miniHOWTO, Token Ring mini-HOWTO, VPN mini-HOWTO, Linux Modem Sharing mini-HOWTO LDP Guides The Linux Installation and Getting Started Guides Chapter 6 covers networking. The major one is the Linux Network Administrators Guide. It is a touch old (March 1996) but it was actually published by O'Rielly and Associates (http://www.ora.com/) but is also freely available as part of the Linux Documentation Project. Linux network project Development on the Linux networking code is an on-going project. The project leader maintains a Web site which contains information about the current developments. It's located at http://www.uk.linux.org/NetNews.html comp.os.linux.networking A newsgroup specifically for discussions about Linux networking. TCP/IP introduction and administration, Documents produced by Rutgers University. Available from ftp://athos.rutgers.edu/runet/ with the filenames tcp-ip-intro and tcp-ip-admin as either Word documents or postscript files. Should also be present on the 85321 website/CD-ROM RFC Database RFCs (Request for comments) are the standards documents for the Internet. A Web-based interface to the collection of RFCs is available from http://pubweb.nexor.co.uk/public/rfc/index/rfc.html Linux for an ISP A number of Internet Service Providers from throughout the world use Linux servers. There is a Web page which maintains a list of of links of interest to these folk. It is available at http://www.anime.net/linuxisp/ Some of the links are dated.

The Overview
This chapter introduces the process and knowledge for connecting a Linux machine to a TCP/IP network. There are many other types of networking protocols but TCP/IP is the protocol family on the Internet so that is the one we concentrate on. Creating a TCP/IP network does not necessarily mean you are connected to the Internet. You can have a TCP/IP network between the two computers you have at home.

What you need

In order to create some sort of TCP/IP network using Linux you will need the following

David Jones (20.01.00)

Page 381

85321, Systems Administration

Chapter 15, Networks: The Connection

Networking hardware You will need to make some sort of connection between the machines on your network so they can communicate. Linux supports a wide range of networking hardware. You can only use networking hardware Linux supports (unless you want to start writing device drivers). Appropriately configured kernel To use your network hardware the kernel must contain the appropriate device driver or have access to an appropriate module. The kernel also requires a number of other components which provide necessary low-level support for networking. If you are using some sort of strange hardware you may need to recompile the kernel to include support for your hardware. Network configuration tools These should be already present on most Linux systems and are used to configure networking. Network applications These are the topic of the next chapter and again most are supplied with the common Linux distributions. These provide the higher level services such as email, Web and file sharing. Network information This information is necessary to configure your system on the network. It includes your machines IP address, the network address, the broadcast and netmask addresses, the router address and the address of your DNS server.

What you do
To install your Linux box onto a network you move on up the layers with steps something like the following Obtain the appropriate hardware Connect it to your system Configure your kernel to recognise the hardware Configure the network software Test the connection

TCP/IP Basics
Before going any further it is necessary to introduce some of the basic concepts related to TCP/IP networks. An understanding of these concepts is essential for the the next steps in connecting a Linux machine to a network. If you find the following too confusing or disjointed please refer to some of the other resources mentioned at the start of this chapter. The concepts introduced in the following includes hostnames Every machine (also known as a host) on the Internet has a name. This section introduces hostnames and related concepts.

David Jones (20.01.00)

Page 382

85321, Systems Administration

Chapter 15, Networks: The Connection

IP addresses Each network interface on the network also has a unique IP address. This section discusses IP addresses, the components of an IP address, subnets, network classes and other related issues. Name resolution Human beings use hostnames while the IP protocols use IP addresses. There must be a wy, name resolution, to convert hostnames into IP addresses. This section looks at how this is achieved. Routing When network packets travel from your computer to a Web site in the United States there are normally a multitude of different paths that packet can take. The decisions about which path it takes are performed by a routing algorithm. This section briefly discusses how routing occurs.

Hostnames
Most computers on a TCP/IP network are given a name, usually known as a host name (a computer can be known as a host). The hostname is usually a simple name used to uniquely identify a computer within a given site. A fully qualified Internet host name, also known as a fully qualified domain name (FQDNFQDN), uses the following format
hostname.site.domain.country

hostname A name by which the computer is known. This name must be unique to the site on which the machine is located. site A short name given to the site (company, University, government department etc) on which the machine resides. domain Each site belongs to a specific domain. A domain is used to group sites of similar purpose together. Table 15.? provides an example of some domain names. Strictly speaking a domain name also includes the country. country Specifies the actual country in which the machine resides. Table 15.? provides an example of some country names. You can see a list of the country codes at http://www.bcpl.net/~jspath/isocodes.html

For example the CQU machine jaspers fully qualified name is jasper.cqu.edu.au, where jasper is the hostname, cqu is the site name, the domain is edu and the country is au.

David Jones (20.01.00)

Page 383

85321, Systems Administration

Chapter 15, Networks: The Connection

Domain
edu com gov net

Purpose Educational institution, university or school Commercial company Government department Networking companies
Example Table 15.1 Internet domains

Country code nothing or us

au uk In Ca Fr

Country United States Australia United Kingdom India Canada France

Table 15.2 Example Country Codes

hostname Under Linux the hostname of a machine is set using the hostname command. Only the root user can set the hostname. Any other user can use the hostname command to view the machines current name.
root@faile david]# hostname faile.cqu.edu.au [root@faile david]# hostname fred [root@faile david]# hostname fred

Changes to the hostname performed using the hostname command will not apply after you reboot a RedHat Linux computer. RedHat Linux sets the hostname during startup from one of its configuration files, /etc/sysconfig/network This is the file which is changed by the GUI tools provided with RedHat. If you wish a change in hostname to be retained after you reboot you will have to change this file.

Qualified names
jasper.cqu.edu.au is a fully qualified domain name and uniquely identifies the machine jasper on the CQU campus to the entire Internet. There cannot be another machine called jasper at CQU. However there could be another machine called jasper at James Cook University in Townsville (its fully qualified name would be jasper.jcu.edu.au).

David Jones (20.01.00)

Page 384

85321, Systems Administration

Chapter 15, Networks: The Connection

A fully qualified name must be unique to the entire Internet. Which implies every hostname on a site should be unique. Not qualified It is not always necessary to specify a fully qualified name. If a user on aldur.cqu.edu.au enters the command telnet jasper the networking software assumes that because it isnt fully qualified hostname the user means the machine jasper on the current site (cqu.edu.au).

IP/Internet Addresses
Alpha-numeric names, like hostnames, cannot be handled efficiently by computers, at least not as efficiently as numbers. For this reason, hostnames are only used for us humans. The computers and other equipment involved in TCP/IP networks use numbers to identify hosts on the Internet. These numbers are called IP addresses. This is because it is the Internet Protocol (IP) which provides the addressing scheme. IP addresses are currently 32 bit numbers, IPv6 the next generation of IP uses 128 bit address. IP addresses are usually written as four numbers separated by full stops (called dotted decimal form) e.g. 132.22.42.1. Since IP addresses are 32 bit numbers, each of the numbers in the dotted decimal form are restricted to between 0-255 (32 bits divide by 4 numbers gives 8 bits per number and 255 is the biggest number you can represent using 8 bits). This means that 257.33.33.22 is an invalid address. Dotted Quad to Binary The address 132.22.42.1 in dotted decimal form is actually stored on the computer as 10000100 00010110 00101010 00000001. Each of the four decimal numbers represent one byte of the final binary number 132 = 10000100 22 = 00010110 42 = 00101010 1 = 00000001

The convervsion from dotted quad to binary (and back again) is important for some of the following concepts. Networks and hosts An IP address actually consists of two parts a network portion, and This is used to identify the network that the machine belongs to. Hosts on the same network will have this portion of the IP address in common. This is one of the reasons why IP masquerading is reqired for mobile computers (e.g. laptops). If you move a computer to a different network you must

David Jones (20.01.00)

Page 385

85321, Systems Administration

Chapter 15, Networks: The Connection

give it a different IP address which includes the network address of the new network it is connected to. the host portion. This is the part which uniquely identifies the host on the network.

The network portion of the address forms the high part of the address (the bit that appears on the left hand side of the number). The size of the network and host portions of an IP address is specified by another 32 bit number called the netmask (also known as the subnet mask)netmask. To calculate which part of an IP address is the network and which the host the IP address and the subnet mask are treated as binary numbers (see diagram 15.?). Each bit of the subnet mask and the IP address are compared and if the bit is set in both the IP address and the subnet mask then the bit is set in the network address, if the bit is set in the IP address but not set in the subnet mask then the bit is set in the host address.

For example
IP address netmask 138.77.37.21 255.255.255.0 10001010 01001101 00100101 00100101 11111111 11111111 11111111 00000000

network address 138.77.37.0 10001010 01001101 00100101 00000000 host address 0.0.0.21 00000000 00000000 00000000 00100101

The Internet is a network of networks

The structure of IP addresses can give you some idea of how the Internet works. It is a network of networks. You start with a collection of machines all connected via the same networking hardware, a local area network. All the machines on this local area network will have the same network address, each machine also has a unique host address. The Internet is formed by connecting a lot of local area networks together. For example In Figure 15.5 there are two networks, 138.77.37.0 and 138.77.36.0. These are two networks on the Rockhampton campus of Central Queensland University and both use ethernet as their networking hardware. This means that when a computer on the 37 subnet (the network with the network address 138.77.37.0) wants to send information to another computing on the 37 subnet it simply uses the characteristics of ethernet. The information is placed on the ethernet network and gets broadcasted to every ethernet card on the network. The ethernet card which has the appropriate address is the only one which accepts the information. However, if the machine 138.77.37.37 wants to send information to the machine 138.77.36.15 it's a bit more complex. Since both computers are on separate networks the machine 138.77.37.37 just can't send information to the machine 138.77.36.15. Instead it has to use a gateway machine (only
David Jones (20.01.00) Page 386

85321, Systems Administration

Chapter 15, Networks: The Connection

rarely is the gateway machine a computer but it can be). The gateway machine actually has two network connections. One connection to the 138.77.37.0 network and the other to the 138.77.36.0 network. It is via this dual connection that the gateway acts as the connection between the two networks. The gateway knows that it should grab any and all packets on the 138.77.36.0 network destined for the 138.77.37.0 network (and vice versa). When it grabs these packets the gateway machine transfers them from the network device connected to the sending network to the network device connected to the receiving network.

Figure 15.5 A simple gateway

This process is repeated for other networks. Each network is then connected to each other via devices called routers, or perhaps gateways. This is a very simple example. Assigning IP addresses Some IP addresses are reserved for specific purposes and you should not assign these addresses to a machine. Table 15.3 lists some of these addresses Address xx.xx.xx.0 xx.xx.xx.1 xx.xx.xx.255 127.0.0.1
*

Purpose network address gateway address * broadcast address loopback address

set standard

this is not a

Table 15.3 Reserved IP addresses

David Jones (20.01.00)

Page 387

85321, Systems Administration

Chapter 15, Networks: The Connection

As mentioned above 127.0.0.1 is a special IP address. It refers to the local host (or the loopback address). The local host allows software to address the local machine in exactly the same way it would address a remote machine. For those of you without network connections the localhost will be the only method you can use to experiment with the concepts introduced in this and the following chapter. As shown in the previous examples gateways and routers are able to distribute data from one network to another because they are actually physically connected to two or more networks through a number of network interfaces. Figure 15.5 provides a representation of this. The machine in the middle, the gateway machine, has two network interfaces. One has the IP address 138.77.37.1 and the other 138.77.36.1 (its common practice for a networks gateway machine to have the host id 1, but by no means compulsory). By convention the network address is the IP address with a host address that is all 0s. The network address is used to identify a network. The broadcast address is the IP address with the host address set to all 1s and is used to send information to all the computers on a network, typically used for routing and error information. Network Classes During the development of the TCP/IP protocol stack IP addresses were divided into classes. There are three main address classes, A, B and C. Table 15.? summarises the differences between the three classes. The class of an IP address can be deduced by the value of the first byte of the address. Class A B C Multicast First byte value 1 to 126 128 to 191 192 to 223 224 239 Netmask 255.0.0.0 255.255.0.0 255.255.255.0 240.0.0.0
Table 15.4 Network classes

Number of hosts 16 million 64,000 254

If you plan on setting up a network that is connected to the Interet the addresses for your network must be allocated to you by central controlling organisation. You can't just choose any set of addresses you wish, chances are they are already taken my some other site. If your network will not be connected to the Internet you can choose from a range of addresses which have been set aside for this purpose. These addresses are shown in Table 15.4

David Jones (20.01.00)

Page 388

85321, Systems Administration

Chapter 15, Networks: The Connection

Network class Addresses A B C 10.0.0.0 to 10.255.255.255 172.16.0.0 to 172.31.255.255 192.168.0.0 to 192.168.255.255

networks

Table 15.5 Networks reserved for private

Subnets Central Queensland University has a class B network address, 138.77.0.0. This would imply that you could make the following assumptions about the IP address 138.77.1.1. The network address is 138.77.0.0 and that the host address is 1.1, this is after all how a class B address is defined. If you did make these implications you would be wrong. CQU has decided to break its available IP addresses into further networks, called subnets. Subnetting works by moving the dividing line between the network address bits and the host address bits. Instead of using the first two bytes for the network address CQU uses subnetting to use the first three bytes. This is achieved by setting the netmask to 255.255.255.0. This means that the address 138.77.1.1 actually breaks up into a network address 138.77.1.0 and a host address of 1. The network 138.77.1.0 is said to be a subnet of the larger 138.77.0.0 network. Why subnet? Subnetting is used for a number of reasons including security reasons, Using ethernet all hosts on the same network can see all the packets on the network. So it makes sense to put the computers in student labs on a different network to the computer on which student results are placed. physical reasons, Networking hardware, like ethernet, has physical limitations. You cant put machines on the Mackay campus on the same network as machines on the Rockhampton campus (they are separated by about 300 kilometers). political reasons, and There may be departments or groups within an organisation that have unique needs or want to control their own network. This can be achieved by subnetting and allocating them their own network. hardware and software differences. Someone may wish to use completely different networking hardware and software.

David Jones (20.01.00)

Page 389

85321, Systems Administration

Chapter 15, Networks: The Connection

"Strange" subnets Generally subnet masks are byte oriented, for example 255.255.255.0. This means that divide between the network portion of the address and the host portion occurs on a byte boundary. However it is possible and sometimes necessary to use bit oriented subnet masks, for example 255.255.255.224. Bit oriented implies that this division occurs within a byte. For example a small company with a class C Internet address might use the subnet mask 255.255.255.224.

Exercises
15.1.

Complete the following table by calculating the network and host addresses. (refer back to the example earlier in the chapter)

IP address
178.86.11.1 230.167.16.132 132.95.132.5

Subnet mask
255.255.255.0 255.255.255.192 255.255.240.0

Network address

Host address

Name resolution
We have a problem. People will use hostnames to identify individual computers on the network while the computers use the IP address. How are the two reconciled. When you enter http://www.lycos.com/ on your WWW browser the first thing the networking software must do is find the IP address for www.lycos.com. Once it has the IP address it can connect to that machine and download the WWW pages. The process of taking a hostname and finding the IP address is called name resolution. Methods of name resolution There are two methods that can be used to perform name resolution the /etc/hosts file, and the Domain Name Service.

/etc/hosts One way of performing name resolution is to maintain a file that contains a list of hostnames and their equivalent IP addresses. Then when you want to know a machines IP address you look up the file.

David Jones (20.01.00)

Page 390

85321, Systems Administration

Chapter 15, Networks: The Connection

Under UNIX the file is /etc/hosts. /etc/hosts is a text file with one line per host. Each line has the format
IP_address hostname aliases

Comments can be indicated by using the hash # symbol. Aliases are used to indicate shorter names or other names used to refer to the same host. For example For example the hosts file of the machine aldur looks like this
# every machine 127.0.0.1 138.77.36.29 138.77.1.1 138.77.37.28 has the localhost localhost aldur.cqu.edu.au jasper.cqu.edu.au pol.cqu.edu.au entry loopback aldur jasper pol

Problems with /etc/hosts When a user on aldur enters the command telnet jasper.cqu.edu.au the software first looks in the hosts file for an entry for jasper. If it finds an entry it obtains jaspers IP address and then can execute the command. What happens if the user enters the command telnet knuth. There isnt an entry for knuth in the hosts file. This means the IP address of knuth cant be found and so the command cant succeed. One solution would be to add an entry in the hosts file for every machine the users of aldur wish to access. With over two million machines on the Internet it should be obvious that this is not a smart solution. Domain name service (DNS) The following reading on the DNS was taken from
http://www.aunic.net/dns.html

In the early days of the Internet, all host names and their associated IP addresses were recorded in a single file called hosts.txt, maintained by the Network Information Centre in the USA. Not surprisingly, as the Internet grew so did this file, and by the mid-80s it had become impractically large to distribute to all systems over the network, and impossible to keep up to date. The Internet Domain Name System (DNS) was developed as a distributed database to solve this problem. Its primary goal is to allow the allocation of host names to be distributed amongst multiple naming authorities, rather than centralised at a single point. DNS structure The DNS is arranged as a hierarchy, both from the perspective of the structure of the names maintained within the DNS, and in terms of the delegation of naming authorities. At the top of the hierarchy is the root domain "." which is administered by the Internet Assigned Numbers Authority (IANA). Administration of the root domain gives the IANA the authority to allocate domains beneath the root, as shown in the diagram below:
David Jones (20.01.00) Page 391

85321, Systems Administration

Chapter 15, Networks: The Connection

The process of assigning a domain to an organisational entity is called delegating, and involves the administrator of a domain creating a sub-domain and assigning the authority for allocating sub-domains of the new domain the subdomains administrative entity. This is a hierarchical delegation, which commences at the "root" of the Domain Name Space ("."). A fully qualified domain name, is obtained by writing the simple names obtained by tracing the DNS hierarchy from the leaf nodes to the root, from left to right, separating each name with a stop ".", eg.
fred.xxxx.edu.au

is the name of a host system (huxley) within the XXXX University (xxx), an educational (edu) institution within Australia (au). The sub-domains of the root are known as the top-level domains, and include the edu (educational), gov (government), and com (commercial) domains. Although an organisation anywhere in the world can register beneath these three-character top level domains, the vast majority that have are located within, or have parent companies based in, the United States. The top-level domains represented by the ISO two-character country codes are used in most other countries, thus organisations in Australia are registered beneath au. The majority of country domains are sub-divided into organisational-type subdomains. In some countries two character sub-domains are created (eg. ac.nz for New Zealand academic organisations), and in others three character subdomains are used (eg. com.au for Australian commercial organisations). Regardless of the standard adopted each domain may be delegated to a separate authority. Organisations that wish to register a domain name, even if they do not plan to establish an Internet connection in the immediate short term, should contact the administrator of the domain which most closely describes their activities. Even though the DNS supports many levels of sub-domains, delegations should only be made where there is a requirement for an organisation or organisational sub-division to manage their own name space. Any sub-domain administrator must also demonstrate they have the technical competence to operate a domain name server (described below), or arrange for another organisation to do so on their behalf.

David Jones (20.01.00)

Page 392

85321, Systems Administration

Chapter 15, Networks: The Connection

Domain Name Servers The DNS is implemented as collection of inter-communicating nameservers. At any given level of the DNS hierarchy, a nameserver for a domain has knowledge of all the immediate sub-domains of that domain. For each domain there is a primary nameserver, which contains authoritative information regarding Internet entities within that domain. In addition Secondary nameservers can be configured, which periodically download authoritative data from the primary server. Secondary nameservers provide backup to the primary nameserver when it is not operational, and further improve the overall performance of the DNS, since the nameservers of a domain that respond to queries most quickly are used in preference to any others.
/etc/resolv.conf

When performing a name resolution most UNIX machines will check their /etc/hosts first and then check with their name server. How does the machine know where its domain name server is. The answer is in the /etc/resolv.conf file. resolv.conf is a text file with three main types of entries # comments Anything after a # is a comment and ignored. domain name Defines the default domain. This default domain will be appended to any hostname that does not contain a dot. nameserver address This defines the IP address of the machines domain name server. It is possible to have multiple name servers defined and they will be queried in order (useful if one goes down).

For example The /etc/resolv.conf file from my machine is listed below.

domain cqu.edu.au nameserver 138.77.5.6 nameserver 138.77.1.1

Routing
So far weve looked at names and addresses that specify the location of a host on the Internet. We now move onto routing. Routing is the act of deciding how each individual datagram finds its way through the multiple different paths to its destination. Simple routing For most UNIX computers the routing decisions they must make are simple. If the datagram is for a host on the local network then the data is placed on the

David Jones (20.01.00)

Page 393

85321, Systems Administration

Chapter 15, Networks: The Connection

local network and delivered to the destination host. If the destination host is on a remote network then the datagram will be forwarded to the local gateway. The local gateway will then pass it on further. However, a network the size of the Internet cannot be constructed with such a simple approach. There are portions of the Internet where routing is a much more complex business, too complex to be covered as a portion of one week of a third year unit. Routing tables Routing is concerned with finding the right network for a datagram. Once the right network has been found the datagram can be delivered to the host. Most hosts (and gateways) on the Internet maintain a routing table. The entries in the routing table contain the information to know where to send datagrams for a particular network. Constructing the routing table The routing table can be constructed in one of two ways constructed by the Systems Administrator, sometimes referred to as static routes, dynamically created by a number of different available routing protocols

The dynamic creation by routing protocols is complex and beyond the scope of this subject.

Exercises
15.2.

Why is the name server in /etc/resolv.conf specified using an IP address and not a hostname?

TCP/IP Basics Conclusion

The Internet is a network of networks. Each network has its own network address. Each computer on those networks has its own network address. Network addresses are allocated in classes. You cant simply choose an IP address yourself. It must match the network you are connecting to and not be used by anyone else. Most organisations with a range of IP address will split them into subnets. Software and hardware use IP addresses to identify computers. People use hostnames. Name resolution makes the connection between a hostname and an IP address (and vice versa). On a small scale name resolution can be done with a local file. However, scaling to a large network requires the use of the Domain Name Service. Routing is the act of delivering packets of information to the appropriate place. With a single physical network routing is quite straight forward. However with a large network of networks maintaining the rules about the routes from one to another network can get quite complex.

David Jones (20.01.00)

Page 394

85321, Systems Administration

Chapter 15, Networks: The Connection

Network Hardware
The first step in connecting a machine to a network is to find out what sort of network hardware you will be using. The aim of this unit and this chapter is not to give you a detailed introduction to networking hardware. If you are interested in the topic there are a number of readings and resources mentioned throughout this section. Before you can use a paticular type of networking hardware, or any hardware for that matter, there must be support for that device in the Linux kernel. If the kernel doesnt support the required hardware then you cant use it. Currently the Linux kernel offers support for the networking hardware outlined in list below. For more detailed information about hardware support under Linux refer to the Hardware Compatibility HOWTO available from your nearest mirror of the Linux Documentation Project. Some of the hardware supported includes arcnet, ATM http://lrcwww.epfl.ch/linux-atm/, AX25 amateur radio, FDDI, Frame relay, ISDN, modems, serial and parallel, radio modem, token ring, X.25, WaveLan, wireless, card, and ethernet In most "normal" situations the networking hardware being used will be either modem A modem is a serial device so your Linux kernel should support the appropriate serial port you have in your computer. The networking protocol used on a modem will be either SLIP or PPP which must also be supported by the kernel. ethernet Possibly the most common form of networking hardware at the moment. There are a number of different ethernet cards. You will need to make sure that the kernel supports the particular ethernet card you will be using. The Hardware Compatibility HOW-TO and the Ethernet HOWTO cover this information.

Network devices
As mentioned in chapter 10 the only way a program can gain access to a physical device is via a device file. Network hardware is still hardware so it follows that there should be device files for networking hardware. Under other versions of the UNIX operating system this is true. It is not the case under the Linux operating system. Device files for networking hardware are created, as necessary, by the device drivers contained in the Linux kernel (ethernet and others) or by user programs which make network connections (e.g. modems, PPP connections). These device files are not available for other programs to use. This means I cant execute the command
cat < /etc/passwd > /dev/eth0

The only way information can be sent via the network is by going through the kernel.
David Jones (20.01.00) Page 395

85321, Systems Administration

Chapter 15, Networks: The Connection

Remember, the main reason UNIX uses device files is to provide an abstraction which is independent of the actual hardware being used. A network device file must be configured properly before you can use it send and receive information from the network. The process for configuring a network is discussed later in this chapter. The installation process for RedHat will normally perform some network configuration for you. Tofind out what network devices are currently active on your system have a look at the contents of the file /proc/net/dev/
[david@faile]$ cat /proc/net/dev Inter-| Receive | Transmit face |packets errs drop fifo frame|packets errs drop fifo colls carrier lo: 91 0 0 0 0 91 0 0 0 0 0 eth0: 0 0 0 0 0 60 0 0 0 0 60

On this machine there are two active network devices. lo: the loopback device and eth0: an ethernet device file. If a computer has more than one ethernet interface (network devices are usually called network interfaces) you would normally see entries for eth1 eth2 etc. IP aliasing (talked about more later) is the ability for a single ethernet card to have more than one Internet address (often used when a single computer is acting as the Web server for many different sites). The following example shows the contents of the /proc/net/dev file for a machine using IP aliasing. It is not normal for an ethernet card to have multiple IP addresses, normally each ethernet card/interface will have one IP address.
[david@cq-pan ]$ cat /proc/net/dev Inter-| Receive | Transmit face |packets errs drop fifo frame|packets errs drop fifo colls carrier lo: 285968 0 0 0 0 285968 0 0 0 0 0 eth0:61181891 59 59 0 89 77721923 0 0 0 11133617 57 eth0:0: 48849 0 0 0 0 212 0 0 0 0 0 eth0:1: 10894 0 0 0 0 210 0 0 0 0 0 eth0:2: 481325 0 0 0 0 259 0 0 0 0 0 eth0:3: 29178 0 0 0 0 215 0 0 0 0 0

You can see that the device files for an aliased ethernet device uses the format ethX:Y where X is the number for the ethernet card and Y is the number of the aliased device. Since aliased devices use the same ethernet card they must use the same network, after all you cant connect a single ethernet card to two networks.

Ethernet
The following provides some very brief background information on ethernet which will be useful in the rest of the chapter. Refer to the Ethernet HOWTO for more information. Ethernet addresses Every ethernet card has built into it a 48 bit address (called an Ethernet address or a Media Access Control (MAC) address). The high 24 bits of the address are used to assign a unique number to manufacturers of ethernet addresses and the low 24 bits are assigned to individual ethernet cards made by the manufacturer.

David Jones (20.01.00)

Page 396

85321, Systems Administration

Chapter 15, Networks: The Connection

Some example ethernet addresses, you will notice that ethernet addresses are written using 6 tuples of HEX numbers, are listed below
00:00:0C:03:79:2F 00:40:F6:60:4D:A4 00:20:AF:A4:55:87 00:20:AF:A4:55:7B

Notice that the last two ethernet cards were made by the same manufacturer (with the manufacturers number of 00:20:AF). Ethernet is a broadcast medium Every packet, often called an ethernet frame, of information sent on ethernet contains a source and destination MAC address. The packet is placed on a ethernet network and every machine, actually the ethernet card, on the network looks at the packet. If the card recognises the destination MAC as its own it "grabs" the packet and passes it to the Network access layer. It is possible to configure your ethernet card so that it grabs all packets sent on the network. This is how it is possible to "listen in" on other people on a ethernet network. A single ethernet network cannot cover much more than a couple of hundred meters. How far depends on the type of cabling used.

Converting hardware addresses to Internet addresses

The network access layer, the lowest level of the TCP/IP protocol stack is responsible for converting Internet addresses into hardware addresses, such as MAC addresses. This is how TCP/IP can be used over a large number of different networking hardware. As you might have guessed different networking hardware uses different addressing schemes. Address Resolution Protocol The mapping of ethernet addresses into Internet addresses is performed by the Address Resolution Protocol (ARPARP). ARP maintains a table that contains the translation between IP address and ethernet address. When the machine wants to send data to a computer on the local ethernet network the ARP software is asked if it knows about the IP address of the machine (remember the software deals in IP addresses). If the ARP table contains the IP address the ethernet address is returned. If the IP address is not known a packet is broadcast to every host on the local network, the packet contains the required IP address. Every host on the network examines the packet. If the receiving host recognises the IP address as its own, it will send a reply back that contains its ethernet address. This response is then placed into the ARP table of the original machine (so it knows it next time). The ARP table will only contain ethernet addresses for machines on the local network. Delivery of information to machines not on the local network requires the intervention of routing software which is introduced later in the chapter.

David Jones (20.01.00)

Page 397

85321, Systems Administration

Chapter 15, Networks: The Connection

arp

On a UNIX machine you can view the contents of the ARP table using the arp command. arp -a will display the entire table. The following example shows how the arp cache for a computer is built as it goes. In the first use of the arp command you can see three machines in the cache, centaurus, draal and a ?. The ? is almost certainly one of the NT computers in the student labs at CQU. Draal is one of the Linux computers used by project students and centaurus is the gateway between the 138.77.37 network and the rest of the world.
[root@cq-pan logs]# /sbin/arp a centaurus.cqu.EDU.AU (138.77.37.1) at AA:00:04:00:0B:1C [ether] on eth0 draal.cqu.EDU.AU (138.77.37.100) at 00:20:AF:33:B5:BE [ether] on eth0 ? (138.77.37.46) at <incomplete> on eth0

To see how new entries are added to the cache the next example shows the ping command. ping is often used to test a network connection and to see if a particular machine is alive. In this case Im pinging pug, who also happens to be on the 138.77.37 network.
[root@cq-pan logs]# ping pug PING pug.cqu.edu.au (138.77.37.102): 56 data bytes 64 bytes from 138.77.37.102: icmp_seq=0 ttl=64 time=19.0 ms --- pug.cqu.edu.au ping statistics --1 packets transmitted, 1 packets received, 0% packet loss round-trip min/avg/max = 19.0/19.0/19.0 ms

Since weve now contacted pug and pug is on the same network as this machine its entry should now appear in the arp cache.
[root@cq-pan logs]# /sbin/arp a centaurus.cqu.EDU.AU (138.77.37.1) at AA:00:04:00:0B:1C [ether] on eth0 draal.cqu.EDU.AU (138.77.37.100) at 00:20:AF:33:B5:BE [ether] on eth0 pug.cqu.EDU.AU (138.77.37.102) at 00:20:AF:A4:3B:0F [ether] on eth0 ? (138.77.37.46) at <incomplete> on eth0

There (s)he blows. If pug was not on the same local area network its ethernet address would not be added to the arp cache. Remember, ethernet addresses are only used to communicate with machines on the same ethernet network. For example, if I ping the machine www.cqu.edu.au it wont be added to the arp cache since it is on a different network.
[root@cq-pan logs]# ping www PING plato.cqu.edu.au (138.77.5.4): 56 data bytes 64 bytes from 138.77.5.4: icmp_seq=0 ttl=63 time=1.7 ms --- plato.cqu.edu.au ping statistics --1 packets transmitted, 1 packets received, 0% packet loss round-trip min/avg/max = 1.7/1.7/1.7 ms

SLIP, PPP and point to point

SLIP and PPP, used to connect machines via serial lines (and modems) are not broadcast media. They are simple "point-to-point" connections between two computers. This means that when information is placed on a SLIP/PPP connection only the two computers at either end of that connection can see the

David Jones (20.01.00)

Page 398

85321, Systems Administration

Chapter 15, Networks: The Connection

information. SLIP/PPP are usually used when a computer is connected to a network via a modem or a serial connection. This chapter does not provide any more discussion of SLIP/PPP. However all the basic concepts and the fundamental process for connecting a machine to the network are the same for SLIP/PPP as they are for ethernet. This is one of the advantages of TCP/IP networking being layered. Above a certain level, i.e. when the network interface is configured, the system works the same regardless of the hardware. Refer to the appropriate HOWTOs for more information.

Kernel support for networking

Ensuring that the kernel includes support for your networking hardware is only the first step. In order to supply certain network services it is necessary for them to be compiled into the kernel. The following is a list of some of the services that the Linux kernel can support. The list itself may be a touch out of date. If in doubt refer to the appropriate networking and kernel HOWTOs. IP accounting IP accounting must be compiled into the kernel and is configured with the ipfwadm command. IP accounting allows you to track the number of bytes and packets transmitted over the network connection. This is useful in situations where you must track the network usage of your users. For example, if you are a Internet Service Provider. IP aliasing Essentially, IP aliasing allows your computer to pretend it is more than one computer. In a normal configuration each network device is allocated a single IP address. However there are times when you wish to allocate multiple IP addresses to a computer with a single network interface. The most common example of this is web sites, for example, the websites http://cq-pan.cqu.edu.au/, http://webclass.cqu.edu.au/, and http://webfuse.cqu.edu.au/ are all hosted by one computer. This computer only has one ethernet card and uses IP aliasing to create aliases for the ethernet card. The ethernet cards real IP address is 138.77.37.37 and its three alias addresses are 138.77.37.36, 138.77.37.59 and 138.77.37.108. Normally the interface would only grab the network packets addressed to 138.77.37.37 but with network aliasing it will grab the packets for all three addresses. You can see this in action by using the arp command. Have a look at the hardware addresses for the computers cq-pan, webclass and webfuse. What can you tell?
[david@draal david]$ /sbin/arp Address HWtype HWaddress centaurus.cqu.EDU.AU ether AA:00:04:00:0B:1C webfuse.cqu.EDU.AU ether 00:60:97:3A:AA:85 cq-pan.cqu.EDU.AU ether 00:60:97:3A:AA:85 science.cqu.EDU.AU ether 00:00:F8:01:9E:DA Flags Mask C C C C Iface eth0 eth0 eth0 eth0

David Jones (20.01.00)

Page 399

85321, Systems Administration

Chapter 15, Networks: The Connection

borric.cqu.EDU.AU webclass.cqu.EDU.AU 138.77.37.46

ether ether

00:20:AF:A4:39:39 00:60:97:3A:AA:85 (incomplete)

C C

eth0 eth0 eth0

IP firewall This option allows you to use a Linux computer to implement a firewall. A firewall works by allowing you to selectively ignore certain types of network connections. By doing this you can restrict what access there is to your computer (or the network behind it) and as a result help increase security. The firewall option is closely related to IP accounting, for example it is configured with the same command, ipfwadm. Firewall support has changed in the newer 2.2 kernels. Please refer to the appropriate HOWTOs.

IP encapsulation IP encapsulation is where the IP packet from your machine is wrapped inside another IP packet. This is of particular use mobile IP, IP multicast and the new buzzword Virtual Private Networks (VPNs). IPX IPX protocol is used in Novel Netware systems. Including IPX support in the Linux kernel allows a Linux computer to communicate with Netware machines. IPv6 IPv6, version 6 of the IP protocol, is the next generation of which is slowly being adopted. IPv6 includes support for the current IP protocol. Linux support for IPv6 is slowly developing. You can find more information at http://www.v6.linux.or.jp/ IP masquerade IP masquerade allows multiple computers to use a single IP address. One situation where this can be useful is when you have a single dialup connection to the Internet via an Internet Service Provider (ISP). Normally,such a dialup connection can only be used by the machine which is connected. Even if the dialup machine is on a LAN with other machines connected they cannot access the Internet. However with IP masquerading it is possible to allow all the machines on that LAN access the Internet. Network Address Translation Support for network address translation for Linux is still at an apha stage. Network address translation is the "next version" of IP masquerade. See http://linas.org/linux/load.html for more information. Mobile IP Since an IP address consists of both a network address and a host address it can normally only be used when a machine is connected to the network specified by the network address. Mobile IP allows a machine to be moved to other networks but still retain the same IP. IP encapsulation is used to send packets destined for the mobile machine to its new location. See for

David Jones (20.01.00)

Page 400

85321, Systems Administration

Chapter 15, Networks: The Connection

http://www.hpl.hp.com/personal/Jean_Tourrilhes/Mob ileIP/mip.html more information. IP multicast IP multicast is used to send packets simultaneously to computers and separate IP networks. It is used for a variety of audio and video transmission. See http://www.teksouth.com/linux/multicast/ for more information. EQL EQL allows you to treat multiple point-to-point connections (SLIP, PPP) as a single logical TCP/IP connection.

By default the Linux kernel will have most of the networking services you require already compiled into it. However, if you want to make use of some of the additional features you need to check and possibly recompile the kernel with support for the required feature.

Configuring the connection

Having reached this stage it is assumed that you have connected (or inserted) your networking hardware (in)to your computer and have (if necessary) recompiled the kernel to provide the necessary networking support. This section provides an overview of configuring a network connection for a fairly typical local area network using ethernet. The steps are much the same for other types of hardare.

The Configuration Process

The configuration process includes the following steps Configure the devices Done either at system startup time (ethernet and other permanent connections) or by a user program (on-demand connections such as PPP over modems) this process configures the network devices with the appropriate information including IP address, network address etc. Configure the name resolver This step sets up the DNS so that your system can translate IP addresses into hostnames and vice versa Configure routing Informs the system how it is meant to send information from one network to another.

The following discusses how these steps are performed. If you were planning to include some of the more advanced features available in the Linux kernel, such as IP masquerading, IP aliasing or EQL, you would have to perform additional steps as outlined in the appropriate HOWTOs or manual pages.

David Jones (20.01.00)

Page 401

85321, Systems Administration

Chapter 15, Networks: The Connection

Configuration Related Tools and Files

Configuring the network interface generally makes use of the following tools/files command line tools such as ifconfig The standard tools used to perform the configuration you might use this from the command line or it might be performed by systems startup scripts. device configuration files Most distributions of Linux use text-based files to store the configuration information used by ifconfig. RedHat uses the /etc/sysconfig/network file and the /etc/sysconfig/networkscripts directory to contain these files which are used as the system starts up. The detail in these files includes IP address, netmask and broadcast addresses for the various devices. Network configuration files These files provide details for other network services such as DNS and routing. Startup files Chances are you will want your network to be automatically configured whenever the computer is turned on. This is where the system startup files enter the picture. GUI configuration tools To help ease the load most systems provide some sort of GUI tool which allows you to perform many of these tasks.

Configuring the device/interface

Earlier in the chapter the concept of a network device was introduced. The following section examines what is required to configure the network device so that it operates. Configuring the network device draws on some of the basic TCP/IP concepts introduced in previous sections. The loopback device/interface The loopback device is a special case. It is always present and is used to provide access to your own machine. Even if you do not have a network connection you will be able to use the loopback interface to test some of the basic networking services. The loopback interface always has the IP address 127.0.0.1. Whenever you use the IP address 127.0.0.1 you are connecting to your own computer. ifconfig Network interfaces are configured using the ifconfig command and has the standard format for turning a device on
ifconfig device_name IP_address netmask netmask up

For example
David Jones (20.01.00) Page 402

85321, Systems Administration

Chapter 15, Networks: The Connection

ifconfig eth0 138.77.37.26 netmask 255.255.255.0 up

Configures the first ethernet address with the IP address of 138.77.37.26 and the netmask of 255.255.255.0.
ifconfig lo 127.0.0.1

Configures the loopback address appropriately. Other parameters for the ifconfig command include
up and down These parameters are used to take the device up and down (turn it on and off). ifconfig eth0 down will disable the eth0 interface and will require an ifconfig command like the first example above to turn it back on. -arp

Will turn on/off the address resolution protocol for the specified interface.
-pointtopoint addr

Used to specify the IP address (addr) of the computer at the far end of a point to point link. RedHat Configuration Files Redhat uses the /etc/sysconfig/network file and the contents of the /etc/sysconfig/network-scripts directory to help in the configuration of network devices during the startup of the system. Hopefully you can draw some useful conclusions from the following examples
[root@cq-pan sysconfig]# cat network NETWORKING=yes FORWARD_IPV4=false HOSTNAME=cq-pan.cqu.edu.au DOMAINNAME=cqu.edu.au GATEWAY=138.77.37.1 GATEWAYDEV=eth0 [root@cq-pan sysconfig]# cat network-scripts/ifcfg-eth0 DEVICE=eth0 IPADDR=138.77.37.37 NETMASK=255.255.255.0 NETWORK=138.77.37.0 BROADCAST=138.77.37.255 ONBOOT=yes

The GUI configuration tools for RedHat Linux modify these files. The script which actually starts networking on a RedHat Linux machine is /etc/rc.d/init.d/network A more indepth explanation of these files please refer to the RedHat manuals for 6.1.

Configuring the name resolver

Once the device/interface is configured you can start using the network. However youll only be able to use IP addresses. At this stage the networking system on your computer will not know how to resolve hostnames (convert hostnames into IP addresses). So if I was configuring a machine on the

David Jones (20.01.00)

Page 403

85321, Systems Administration

Chapter 15, Networks: The Connection

138.77.37 subnet (this is the student subnet in the IT building) at CQU I would be able to execute commands like
telnet 138.77.37.37

but I would not be able to execute commands such as

telnet cq-pan.cqu.edu.au

Even though the IP address for the machine cq-pan.cqu.edu.au is 138.77.37.37 the networking on my machine doesnt know how to do the translation. This is where the name resolver and its associated configuration files enter the picture. In particular the three files well be looking at are /etc/resolv.conf Specifies where the main domain name server is located for your machine. /etc/hosts.conf Allows you to specify how the name resolver will operate. For example, will it ask the domain name server first or look at a local file. /etc/hosts A local file which specifies the IP/hostname association between common or local computers.

The following is an excerpt from the NET-3 HOW-TO which describes these files in a bit more detail. /etc/resolv.conf The /etc/resolv.conf is the main configuration file for the name resolver code. Its format is quite simple. It is a text file with one keyword per line. There are three keywords typically used, they are: domain This keyword specifies the local domain name. search This keyword specifies a list of alternate domain names to search for a hostname nameserver This keyword, which may be used many times, specifies an IP address of a domain name server to query when resolving names

An example /etc/resolv.conf might look something like:

domain maths.wu.edu.au search maths.wu.edu.au wu.edu.au nameserver 192.168.10.1 nameserver 192.168.12.1

This example specifies that the default domain name to append to unqualified names (ie hostnames supplied without a domain) is maths.wu.edu.au and that if the host is not found in that domain to also try the wu.edu.au domain directly. Two nameservers entry are supplied, each of which may be called upon by the name resolver code to resolve the name.

David Jones (20.01.00)

Page 404

85321, Systems Administration

Chapter 15, Networks: The Connection

/etc/host.conf

The /etc/host.conf file is where you configure some items that govern the behaviour of the name resolver code. The format of this file is described in detail in the resolv+ man page. In nearly all circumstances the following example will work for you:
order hosts,bind

multi on This configuration tells the name resolver to check the /etc/hosts file before attempting to query a nameserver and to return all valid addresses for a host found in the /etc/hosts file instead of just the first.
/etc/hosts

The /etc/hosts file is where you put the name and IP address of local hosts. If you place a host in this file then you do not need to query the domain name server to get its IP Address. The disadvantage of doing this is that you must keep this file up to date yourself if the IP address for that host changes. In a well managed system the only hostnames that usually appear in this file are an entry for the loopback interface and the local hosts name.
# /etc/hosts 127.0.0.1 192.168.0.1 localhost loopback this.host.name

You may specify more than one host name per line as demonstrated by the first entry, which is a standard entry for the loopback interface.

Configuring routing
Having performed each of the preceding steps the networking on your computer will still not be working 100% correctly. For example, assume Im adding a machine to the 138.77.37 subnet at CQU with the IP address as 138.77.37.105 and the hostname fred. Ive configured the network interface and set up the following files (For the following discussion it is important to realise that CQU has a class B address, 138.77, and creates subnets which look like class C address, i.e. 138.77.37, 138.77.1 and 138.77.5 are all separate subnets)
/etc/resolv.conf
search cqu.edu.au nameserver 138.77.5.6 nameserver 138.77.1.23

/etc/host.conf

order hosts,bind multi on

/etc/hosts
localhost.localdomain fred.cqu.edu.au cq-pan.cqu.edu.au

127.0.0.1 localhost 138.77.37.105 fred 138.77.37.37 cq-pan

David Jones (20.01.00)

Page 405

85321, Systems Administration

Chapter 15, Networks: The Connection

Now, see what happens when I execute the following commands

[david@fred david]$ ping cq-pan.cqu.edu.au PING cq-pan.cqu.edu.au (138.77.37.37): 56 data bytes 64 bytes from 138.77.37.37: icmp_seq=0 ttl=63 time=1.1 ms 64 bytes from 138.77.37.37: icmp_seq=1 ttl=63 time=1.0 ms 64 bytes from 138.77.37.37: icmp_seq=2 ttl=63 time=1.0 ms --- cq-pan.cqu.edu.au ping statistics --3 packets transmitted, 3 packets received, 0% packet loss round-trip min/avg/max = 1.0/1.0/1.1 ms [root@fred network-scripts]# ping jasper.cqu.edu.au ping: unknown host jasper.cqu.edu.au

Why the difference? Weve setup the name resolution configuration files properly so why cant it resolve the name jasper.cqu.edu.au to the IP address 138.77.1.1? Have a look at the IP addresses of the domain name servers specified in the /etc/resolv.conf file above? What can you tell about these hosts? The major difference between the domain name servers and our new host fred is that they are on separate subnets. At this stage our host has not been told how it is meant to send information from its own subnet to other subnets (remember the discussion earlier in the chapter about arp and ethernet being a broadcast medium?). fred.cqu.edu.au is able to use the cq-pan.cqu.edu.au hostname because it is specified in the /etc/hosts file and it can send information to that machine because it is on the same subnet. Because the domain name servers are on another subnet the networking software on the machine doesnt know how to communicate with them. An example of what happens can be seen in the following command where rather than use jasper.cqu.edu.aus hostname we use the IP address.
[david@fred david]$ ping 138.77.1.1 PING 138.77.1.1 (138.77.1.1): 56 data bytes ping: sendto: Network is unreachable ping: wrote 138.77.1.1 64 chars, ret=-1 ping: sendto: Network is unreachable ping: wrote 138.77.1.1 64 chars, ret=-1 --- 138.77.1.1 ping statistics --2 packets transmitted, 0 packets received, 100% packet loss

The solution to this problem is to configuring the routing software on our computer. Routing is the art of deciding how to send IP packets from one host to another, particularly where there are possibly multiple paths that could be used. In our example above we have to specify how the networking software is to deliver IP packets from our current subnet, 138.77.37, to other subnets. Routing is a huge and complex topic. It is not possible to provide a detailed introduction in the confines of this text. If you need more information you should take a look at the relevant HOWTOs and especially the Linux Network Administrators Guide.

David Jones (20.01.00)

Page 406

85321, Systems Administration

Chapter 15, Networks: The Connection

An explanation of routing tables and commands The following is an excerpt rom the NET-3 HOW-TO which briefly describes the routing table and the commands used to manipulate it. Ok, so how does routing work ? Each host keeps a special list of routing rules, called a routing table. This table contains rows which typically contain at least three fields, the first is a destination address, the second is the name of the interface to which the datagram is to be routed and the third is optionally the IP address of another machine which will carry the datagram on its next step through the network. In linux you can see this table by using the following command:
# cat /proc/net/route

or by using either of the following commands:

# /sbin/route n # /bin/netstat -r

The routing process is fairly simple: an incoming datagram is received, the destination address (who it is for) is examined and compared with each entry in the table. The entry that best matches that address is selected and the datagram is forwarded to the specified interface. If the gateway field is filled then the datagram is forwarded to that host via the specified interface, otherwise the destination address is assumed to be on the network supported by the interface. To manipulate this table a special command is used. This command takes command line arguments and converts them into kernel system calls that request the kernel to add, delete or modify entries in the routing table. The command is called route. A simple example. Imagine you have an ethernet network. Youve been told it is a class-C network with an address of 192.168.1.0. Youve been supplied with an IP address of 192.168.1.10 for your use and have been told that 192.168.1.1 is a router connected to the Internet. The first step is to configure the interface as described earlier. You would use a command like:
# ifconfig eth0 192.168.1.10 netmask 255.255.255.0 up

You now need to add an entry into the routing table to tell the kernel that datagrams for all hosts with addresses that match 192.168.1.* should be sent to the ethernet device. You would use a command similar to:
# route add -net 192.168.1.0 netmask 255.255.255.0 eth0

Note the use of the -net argument to tell the route program that this entry is a network route. Your other choice here is a -host route which is a route that is specific to one IP address. This route will enable you to establish IP connections with all of the hosts on your ethernet segment. But what about all of the IP hosts that arent on your ethernet segment ?

David Jones (20.01.00)

Page 407

85321, Systems Administration

Chapter 15, Networks: The Connection

It would be a very difficult job to have to add routes to every possible destination network, so there is a special trick that is used to simplify this task. The trick is called the default route. The default route matches every possible destination, but poorly, so that if any other entry exists that matches the required address it will be used instead of the default route. The idea of the default route is simply to enable you to say "and everything else should go here". In the example Ive contrived you would use an entry like:
# route add default gw 192.168.1.1 eth0

The gw argument tells the route command that the next argument is the IP address, or name, of a gateway or router machine which all datagrams matching this entry should be directed to for further routing. So, your complete configuration would look like:
# ifconfig eth0 192.168.1.10 netmask 255.255.255.0 up # route add -net 192.168.1.0 netmask 255.255.255.0 eth0 # route add default gw 192.168.1.1 eth0

These steps are actually performed automatically by the startup files on a properly configured Linux box.

Network management tools

You might ask, "Why the hell are we playing with all these text files and commands? Why cant we just use the nice GUI tools that come with RedHat". The simple answer is that knowing how to use a GUI tool isnt all that difficult, anyone can learn that. Whats important for a computing professional, like a Systems Administrator, to know is what is going on underneath. There will be times when the GUI doesnt work or the problem you have cant be solved with the GUI. It is at times like this that you will need to understand what is going on underneath. Having said that it can be a lot quicker to perform simple tasks using a GUI than with text files and a command line (depending on your personal preference). The following section introduces the GUI tools RedHat provides to manage and configure networking and also looks at a couple of other useful commands UNIX provides.

RedHat GUI Networking Tools

RedHat supplies a number of GUI administration tools which are all launched from the control-panel application by typing control-panel from a shell (you must be running X-Windows as control-panel is an X application). Each of the icons in the control panel window correspond to one of the GUI tools. Holding the mouse over the icon will cause it to display the name of the tool. Of particular interest to this chapter is the network configuration tool which allows you to configure the hosts, name servers, devices and routing for your system. You might also be interested in using linuxconf a popular GUI configuration tool which provides a number of interfaces.

David Jones (20.01.00)

Page 408

85321, Systems Administration

Chapter 15, Networks: The Connection

nslookup

The nslookup command is used to query a name server and is supplied as a debugging tool. It is generally used to determine if the name server is working correctly and for querying information from remote servers. nslookup can be used from either the command line or interactively. Giving nslookup a hostname will result in it asking the current domain name server for the IP address of that machine. nslookup also has an ls command that can be used to view the entire records of the current domain name server. For example
[david@cq-pan:~]$ nslookup Default Server: circus.cqu.edu.au Address: 138.77.5.6 > jasper Server: circus.cqu.edu.au Address: 138.77.5.6 Name: jasper.cqu.edu.au Address: 138.77.1.1 > exit [david@cq-pan:~]$ nslookup jasper Server: circus.cqu.edu.au Address: 138.77.5.6 Name: jasper.cqu.edu.au Address: 138.77.1.1

netstat
The netstat command is used to display the status of network connections to a UNIX machine. One of the functions it can be used for is to display the contents of the kernel routing table by using the -r switch. For example The following examples are from two machines on CQUs Rockhampton campus. The first one is from telnet jasper
[david@cq-pan:~]$ netstat -rn Kernel routing table Destination Gateway Genmask 138.77.37.0 0.0.0.0 255.255.255.0 127.0.0.0 0.0.0.0 255.0.0.0 0.0.0.0 138.77.37.1 0.0.0.0 bash$ netstat -rn Routing tables Destination Gateway Flags 127.0.0.1 127.0.0.1 UH default 138.77.1.11 UG 138.77.32 138.77.1.11 UG 138.77.16 138.77.1.11 UG 138.77.8 138.77.1.11 UG 138.77.80 138.77.1.11 UG Flags U U UG Metric 0 0 0 Ref Use Iface 0 109130 eth0 0 9206 lo 0 2546951 eth0

Refcnt 56 23 0 0 0 0

Use 7804440 1595585 19621 555 385345 0

Interface lo0 ln0 ln0 ln0 ln0 ln0

David Jones (20.01.00)

Page 409

85321, Systems Administration

Chapter 15, Networks: The Connection

138.77.72 138.77.64 138.77.41

138.77.1.11 138.77.1.11 138.77.1.11

UG UG UG

0 0 0

0 0 0

ln0 ln0 ln0

traceroute

For some reason or another, users on one machine cannot connect to another machine or if they can any information transfer between the two machines is either slow or plagued by errors. What do you do? Remember it is not only the machines at the two ends you have to check. If the two machines are on different networks the information will flow through a number of gateways and routers. It might be one of the gateway machines that is causing the problem. The traceroute command provides a way of discovering the path taken by information as it goes from one machine to another and can be used to identify where problems might be occurring. On the Internet that path may not always be the same. For example The following are the results of a number of executions of traceroute from the machine aldur (138.77.36.29). In the first example the machine knuth is on the same network as aldur. This means that the information can get their directly.
bash$ traceroute knuth traceroute to knuth.cqu.edu.au (138.77.36.20), 30 hops max, 40 byte packets 1 knuth.cqu.EDU.AU (138.77.36.20) 2 ms 2 ms 2 ms

jasper is one network away from aldur

bash$ traceroute jasper traceroute to jasper.cqu.edu.au (138.77.1.1), 30 hops max, 40 byte packets 1 centaurus.cqu.EDU.AU (138.77.36.1) 1 ms 1 ms 1 ms 2 jasper.cqu.EDU.AU (138.77.1.1) 2 ms 1 ms 1 ms

A machine still on the CQU site but a little further away

bash$ traceroute jade traceroute to jade.cqu.edu.au (138.77.7.2), 30 hops max, 40 byte packets 1 centaurus.cqu.EDU.AU (138.77.36.1) 1 ms 1 ms 1 ms 2 hercules.cqu.EDU.AU (138.77.5.3) 4 ms 2 ms 12 ms 3 jade.cqu.EDU.AU (138.77.7.2) 3 ms 13 ms 3 ms

A host still in Australia (but a long way from CQU)

bash$ traceroute archie.au traceroute to archie.au (139.130.23.2), 30 hops max, 40 byte packets 1 centaurus.cqu.EDU.AU (138.77.36.1) 1 ms 1 ms 1 ms 2 tucana.cqu.EDU.AU (138.77.5.27) 2 ms 2 ms 2 ms 3 138.77.32.10 (138.77.32.10) 5 ms 5 ms 5 ms 4 qld.gw.au (139.130.60.1) 21 ms 13 ms 51 ms 5 national.gw.au (139.130.48.1) 35 ms 36 ms 40 ms 6 plaza.aarnet.edu.au (139.130.23.2) 38 ms 35 ms 68 ms

David Jones (20.01.00)

Page 410

85321, Systems Administration

Chapter 15, Networks: The Connection

A host in the Eastern United States.

bash$ traceroute sunsite.unc.edu traceroute to knuth.cqu.edu.au (139.130.23.2), 30 hops max, 40 byte packets 1 centaurus.cqu.EDU.AU (138.77.36.1) 1 ms 1 ms 1 ms 2 tucana.cqu.EDU.AU (138.77.5.27) 2 ms 2 ms 3 ms 3 138.77.32.10 (138.77.32.10) 5 ms 5 ms 5 ms 4 qld.gw.au (139.130.60.1) 13 ms 20 ms 13 ms 5 national.gw.au (139.130.48.1) 51 ms 36 ms 36 ms 6 usa.gw.au (139.130.29.5) 37 ms 36 ms 38 ms 7 usa-au.gw.au (203.62.255.1) 233 ms 252 ms 264 ms 8 * * t3-0.enss144.t3.nsf.net (192.203.230.253) 224 ms 9 140.222.8.4 (140.222.8.4) 226 ms 236 ms 258 ms 10 t3-3.cnss25.Chicago.t3.ans.net (140.222.25.4) 272 ms 293 ms 266 ms 11 t3-0.cnss40.Cleveland.t3.ans.net (140.222.40.1) 328 ms 270 ms 300 ms 12 t3-1.cnss48.Hartford.t3.ans.net (140.222.48.2) 325 ms 355 ms 289 ms 13 t3-2.cnss32.New-York.t3.ans.net (140.222.32.3) 284 ms 319 ms 347 ms 14 t3-1.cnss56.Washington-DC.t3.ans.net (140.222.56.2) 352 ms 299 ms 305 ms 15 t3-1.cnss72.Greensboro.t3.ans.net (140.222.72.2) 319 ms 344 ms 310 ms 16 mf-0.cnss75.Greensboro.t3.ans.net (140.222.72.195) 343 ms 320 ms * 17 cnss76.Greensboro.t3.ans.net (192.103.68.6) 338 ms 319 ms 355 ms 18 192.103.68.50 (192.103.68.50) 338 ms 330 ms 330 ms 19 rtp5-gw.ncren.net (128.109.135.254) 357 ms 361 ms * 20 * rtp2-gw.ncren.net (128.109.70.253) 359 ms 334 ms 21 128.109.13.2 (128.109.13.2) 374 ms 411 ms 451 ms 22 * calypso-2.oit.unc.edu (198.86.40.81) 418 ms 415 ms

There are now a number of visual versions of traceroute, http://www.visualroute.com/, is one of them Exercises
15.3.

In the above example examine the times between machines 6 & 7. Why do you think it takes so long to get from machine 6 to machine 7?

Conclusions
Network protocols are known for consisting of a number of layers. Connecting a Linux box to a network also has a number of layers Select the appropriate hardware Compile an appropriate kernel Configure the network devices with appropriate settings Configure the DNS and routing services Configure any additional networking services

As with network protocol layers, the layers in setting up a network connection also hide detail. For example, once you have connected the hardware and recompiled the kernel configuring network devices is very similar regardless of the networking hardware being used.

David Jones (20.01.00)

Page 411

85321, Systems Administration

Chapter 15, Networks: The Connection

Review Questions
15.1 "The Net", a movie with Sandra Bullock, contained images of a few screens with what appeared to be IP addresses. Some of those supposed IP addresses are listed below. Are they IP addresses? If not, why not? 1. 128.234.15 2. 23.75.345.200 3. 75.258.34.164 15.2 What UNIX commands would you use for the following tasks 1. checking a domain name server for the IP address of the machine
www.seven.com.au

2. finding out whether or not your computer can access, via the network, another machine, 3. finding out what machines information passes through as it goes from your machine to www.whitehouse.gov 4. configure a network interface, 5. display the routing table of your UNIX machine, 6. display the ethernet address of your UNIX machine. 15.3 Explain the relevance of each of the following 1. /etc/hosts 2. /etc/resolv.conf 3. /etc/networks 4. /etc/rc.d/rc.inet1 5. a gateway 15.4 It has been suggested that the layering of the network configuration steps means that configuring the network devices for a PPP connection will be similar to that for an ethernet connection. Refer to the appropriate manual pages and HOWTOs and compare the steps involved in making a PPP connection. For example Where is the device configuration information stored? How is the network device configured? Are the any similarities or differences with network configuratio?
Page 412

David Jones (20.01.00)

85321, Systems Administration

Chapter 15, Networks: The Connection

What about DNS and routing configuration, is there any similarities?

15.5 Youve just started administering a new Linux computer and executed the following three ifconfig commands to discover information about the configuration of the network devices on this machine. Using this output answer the following questions (HINT: remember the discussion of IP aliasing earlier in this chapter?) 1. List the network and host portions of the IP address for each of the network devices listed in the output of these commands. What does the output of these commands tell you about the network configuration of these machines? What would the /proc/net/dev file for this system look like? Can you see what is wrong with the configuration of the networking of this system?
[root@cq-pan logs]# /sbin/ifconfig eth0 eth0 Link encap:Ethernet HWaddr 00:60:97:3A:AA:85 inet addr:138.77.37.37 Bcast:138.77.37.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:61183404 errors:59 dropped:59 overruns:0 TX packets:77722967 errors:0 dropped:0 overruns:0 Interrupt:9 Base address:0xff00 [root@cq-pan logs]# /sbin/ifconfig eth0:1 eth0:1 Link encap:Ethernet HWaddr 00:60:97:3A:AA:85 inet addr:138.77.37.59 Bcast:138.77.37.255 Mask:255.255.255.0 UP BROADCAST RUNNING MTU:1500 Metric:1 RX packets:10894 errors:0 dropped:0 overruns:0 TX packets:210 errors:0 dropped:0 overruns:0 [root@cq-pan logs]# /sbin/ifconfig eth0:2 eth0:2 Link encap:Ethernet HWaddr 00:60:97:3A:AA:85 inet addr:138.77.38.60 Bcast:138.77.38.255 Mask:255.255.255.0 UP BROADCAST RUNNING MTU:1500 Metric:1 RX packets:481325 errors:0 dropped:0 overruns:0 TX packets:259 errors:0 dropped:0 overruns:0

David Jones (20.01.00)

Page 413

85321, Systems Administration

Chapter 16: Network Applications

Chapter
Network Applications

Introduction
The previous chapter looked at how you connect a Linux box to a network and provide some basic services. That is not enough information to produce a useful Linux machine. You need to know how to configure and manage the higher level network services which are expected today including print/file sharing, electronic mail, File Transfer Protocol, World-Wide Web and others. Thats where this chapter comes in. It aims to provide an overview of how network applications work,how they operate and how they are configured. There is no way a single chapter can provide this information about all the available network applications. There are hundreds of them and each one can be quite complex. Instead this chapter focuses on the fundamentals, the concepts which are common to all these applications. If you are comfortable with this knowledge then learning how to configure a new application is quite simple. The chapter closes with a detailed look at some specific network services including file/print sharing, messaging (email) and the World-Wide Web.

Other Resources
Other available resources which examine similar material include HOW-TOs Firewall, IPCHAINS, Intranet Server (though it is a little dated), Mail, Mail User, NFS, NIS, Networking Overview (gives a very good overview of topics related to both networking chapters), SMB, VPN, Virtual Services, WWW, Apache SSL PHP/FI Mini HOW-TOs Apache SSL PHP/FI, Automount, Cipe+Masquerading, ISP Connectivity, NFS-Root, NFS-Root-Client, Qmail+MH, Remove Boot, Remote X Apps, Sendmail Address Rewrite, Sendmail+UUCP, Secure POP via SSH. LAME Sections on DNS Configuration, sections on Windows and Mac file and print sharing, NFS section, configuring the Apache Web server, configuring the Squid HTTP caching proxy, Configuring sendmail. Apache website http://www.apache.org/ Samba website http://www.samba.org The RedHat reference and getting started guide has additional information about many of these topics.

David Jones (20.01.00)

Page 414

85321, Systems Administration

Chapter 16: Network Applications

An expanded an up to date list of resources can be found on the 85321 website.

How it all works

So what is the common details about all the network applications. How do they work? This section provides a general answer to these questions. The provision of network services like FTP, telnet, e-mail and others relies on these following components network ports, Network ports are the logical (that means that ports are an imaginary construct which exists only in software) connections through which the information flows into and out of a machine. A single machine can have thousands of programs all sending and receiving information via the network at the same time. The delivery of this information to the right programs is achieved through ports. Generally each program must have its own port. network daemons, Network daemons are the programs running on the network server machines that sit listening at pre-defined ports waiting for connections from other hosts. These daemons wait for a request, perform some action and send a response back to the program that requested the action. The program which requested the action is a network client. network clients, and Users access network services using client programs. Example network client programs include Netscape, Eudora and the ftp command on a UNIX machine. The client programs turn user requests (e.g. typing in the URL http://www.linux.org/ ) into a request which is sent to a network daemon. The requests and responses which flow between network daemons and network clients must take part in some agreed upon format, a network protocol. network protocols. Network protocols specify how the network clients and servers communicate. They define the small "language" which both use for communication.

The following sections of this chapter go into more detail about each of these sections.

Ports
All network protocols, including http ftp SMTP, use either TCP or UDP to deliver information. TCP and UDP are referred to as transport protocols. Each transport protocol has its own characteristics and which one is used depends on the type of communication which occurs. However, one thing is common between both transport protocols. The addresses they use to identify the source (where they are coming from) and the destination (where they are going to). Obviously the first component of the
David Jones (20.01.00) Page 415

85321, Systems Administration

Chapter 16: Network Applications

source/destination address is the IP address, this identifies the computer. The next component is the port number on that computer.. Every TCP or UDP header contains two 16 bit numbers that are used to identify the source port (the port through which the information was sent) and the destination port (the port through which the information must be delivered.) The IP address is stored in the IP header. Since port numbers are 16 bit numbers, there can be approximately 64,000 (216 is about 64,000) different ports. Some of these ports are used for predefined purposes. The ports 0-256 are used by the network servers for well known Internet services (e.g. telnet, FTP, SMTP). Ports in the range from 256-1024 are used for network services that were originally UNIX specific. Network client programs and other programs should use ports above 1024. Table 16.1 lists some of the port numbers for well known services. Port number 20 21 23 25 80 119
ftp telnet SMTP http nntp

Purpose
ftp-data

(mail) (WWW) (network news)

Table 16.1 Reserved Ports

This means that when you look at a TCP/UDP packet and see that it is addressed to port 25 then you can be sure that it is part of an email message being sent to a SMTP server. A packet destined for port 80 is likely to be a request to a Web server.

Reserved ports
So how does the computer know which ports are reserved for special services? On a UNIX computer this is specified by the file /etc/services. Each line in the services file is of the format
service-name port/protocol aliases

Where service-name is the official name for the service, port is the port number that it listens on, protocol is the transport protocol it uses and aliases is a list of alternate names.
/etc/services

The following is an extract from an example /etc/services file. Most files will be the same, or at least very similar.

echo 7/tcp echo 7/udp discard 9/tcp sink null discard 9/udp sink null systat 11/tcp users daytime 13/tcp daytime 13/udp

David Jones (20.01.00)

Page 416

85321, Systems Administration

Chapter 16: Network Applications

ftp-data 20/tcp ftp 21/tcp telnet 23/tcp smtp 25/tcp mail nntp 119/tcp usenet # Network News Transfer ntp 123/tcp # Network Time Protocol

You should be able to match some of the entries in the above example, or in the /etc/services file on your computer, with the entries in Table 16.1. Exercises
16.1.

Examine your /etc/services file and discover the port on which the following protocols are used
http ssh pop3

Look at ports, netstat

The netstat command can be used for a number of purposes including looking at all of the current active network connections. The following is an example of the output that netstat can produce (its been edited to reduce the size).
[david@cq-pan:~]$ netstat -a Active Internet connections (including servers) Proto Recv-Q Send-Q Local Address Foreign Address root tcp 1 7246 cq-pan.cqu.edu.au:www lore.cs.purdue.e:42468 tcp 0 0 cq-pan.cqu.edu.au:www sdlab142.syd.cqu.:1449 tcp 0 0 cq-pan.cqu.edu.au:www dialup102-4-9.swi:1498 tcp 0 22528 cq-pan.cqu.edu.au:www 205.216.78.103:3058 tcp 1 22528 cq-pan.cqu.edu.au:www barney.poly.edu:47547 tcp 0 0 cq-pan.cqu.edu.au:www eda.mdc.net:2395 tcp 0 22528 cq-pan.cqu.edu.au:www eda.mdc.net:2397 tcp 0 0 cq-pan.cqu.edu.au:www cphppp134.cyberne:1657 tcp 0 22528 cq-pan.cqu.edu.au:www port3.southwind.c:1080 tcp 0 9 cq-pan.cqu.edu.:telnet dinbig.cqu.edu.au:1107 tcp 0 0 cq-pan.cqu.edu.au:ftp ppp2-24.INRE.ASU.:1718 (State) User CLOSING root CLOSE root FIN_WAIT2 root CLOSE root CLOSE root CLOSE root CLOSE root FIN_WAIT2 root CLOSE root ESTABLISHED root FIN_WAIT2 root

Explanation Table 16.2 explains each column of the output. Taking the column descriptions from the table, it is possible to make some observations All of the entries, but the last two, are for people accessing this machines (cq-pan.cqu.edu.au) World-Wide Web server. You can say this because of cq-pan.cqu.edu.au:www. This tells us that the port on the local machine is the www port (port 80). In the second last entry, I am telneting to cq-pan from my machine at home. At that stage my machine at home was called dinbig.cqu.edu.au. The telnet client is using port 1107 on dinbig to talk to the telnet daemon. the last entry is someone connecting to CQ-PANs ftp server,

David Jones (20.01.00)

Page 417

85321, Systems Administration

Chapter 16: Network Applications

the connection for the first entry is shut down but not all the data has been sent (this is what the CLOSING state means). This entry, from a machine from Purdue University in the United States, still has 7246 bytes still to be acknowledged Explanation the name of the transport protocol (TCP or UDP) being used the number of bytes not copied to the receiving process the number of bytes not yet acknowledged by the remote host the local hostname (or IP address) and port of the connection the remote hostname (or IP address) and remote port the state of the connection (only used for TCP because UDP doesnt establish a connection), the values are described in the man page some systems display the user that owns the local program serving the connection
Table 16.2 Columns for net

Column name
Proto Recv-Q Send-Q Local Address Foreign Address State

User

stat

Network daemons
The /etc/services file specifies which port a particular protocol will listen on. For example SMTP (Simple Mail Transfer Protocol, the protocol used to transfer mail between different machines on a TCP/IP network) uses port 25. This means that there should be a network daemon that listens for SMTP connections on port 25 and knows what to do with those connections. This begs some questions How do we know which program acts as the network daemon for which protocol? How is that program started?

How network daemons start

There are two methods by which network daemons are started by startup scripts Daemons started in this manner will show up in a ps list of all the current running processes. These daemons are always running, waiting for a connection on the specified port. This means that the daemon is using up system resources (RAM etc) because it is always in existence but it also means that it is very quick to respond when requests arrive for their services.

David Jones (20.01.00)

Page 418

85321, Systems Administration

Chapter 16: Network Applications

by the inetd daemon The inetd daemon listens at a number of ports and when information arrives, it starts the appropriate network daemon for that port. Which daemon, for which port, is specified in the configuration file /etc/inetd.conf.

Starting a network daemon via inetd is usually done when there arent many connections for that daemon. If a network daemon is likely to get a large number of connections (a busy mail or WWW daemon for example) the daemon for that service should be started in the system startup files and always listen on the port. The reason for this is overhead. Using inetd takes longer because for every connection it needs to first create a new process (and weve seen already that creating new processes can be a relatively expensive process). When the daemon is already running and listening to the port it simply starts handling the request. Of course the draw back with starting daemons in the startup scripts is that they are always there consuming RAM and other resources. Even if they arent being used.

inetd
The /etc/inetd.conf file specifies the network daemons that the inetd daemon should execute. The inetd.conf file consists of one line for each network service using the following format (Table 16.3 explains the purpose of each field).
service-name socket-type protocol flags user daemon_program args

Field
service-name socket-type

Purpose The service name, the same as that listed in /etc/services The type of data delivery services used (we dont cover this). Values are generally stream for TCP, dgram for UDP and raw for direct IP the transport protocol used, the name matches that in the /etc/protocols file how inetd is to behave with regards this service (not explained any further) the username to run the daemon as, usually root but there are some exceptions, generally for security reasons the full path to the program to run as the daemon command line arguments to pass to the daemon program
Table 16.3 Fields of /etc/inetd.conf

protocol

flags

user

daemon_program args

The following is an excerpt from the /etc/inetd.conf on a RedHat Linux machine. Notice that some of the entries have been commented out. This

David Jones (20.01.00)

Page 419

85321, Systems Administration

Chapter 16: Network Applications

means that these services will not be operational. No daemon listening for connections means the service wont work.
shell login #exec #comsat talk ntalk stream stream stream dgram dgram dgram tcp tcp tcp udp udp udp nowait nowait nowait wait wait wait root root root root nobody.tty nobody.tty /usr/sbin/tcpd /usr/sbin/tcpd /usr/sbin/tcpd /usr/sbin/tcpd /usr/sbin/tcpd /usr/sbin/tcpd in.rshd in.rlogind in.rexecd in.comsat in.talkd in.ntalkd

How it works
Whenever the machine receives a request on a port (on which the inetd daemon is listening on), the inetd daemon decides which program to execute on the basis of the /etc/inetd.conf file. Exercises
16.2.

top is a UNIX command which will give you a progressive display of the current running processes. Use top to observer what happens when a network daemon is started. For example, start top and then try to telnet or ftp to your machine. Can you see the appropriate daemon start? (Remember you should be able to use the hostname localhost for your own machine even if you are not on a network.) What happens if you change the /etc/inetd.conf file? Does the inetd daemon pick up the change automatically? How would you notify inetd of the change? Note: you WILL have to experiment to find out the answer to this question. It isnt included in the study material. A suggested experiment is the following: try the command telnet localhost, this should cause inetd to do some work; if it works, comment out the entry in the inetd.conf file for the telnet service try the first command again. Does it work? If it does then inetd hasnt seen the change. How do you tell it?

16.3.

Network clients
All of you will have used a number of network client programs. If you are reading this online you may well be using a WWW browser. Its a network client program. Checking your mail makes use of a network client. A network client is simply a program (whether it is text based or a GUI program) that knows how to connect to a network daemon, pass requests to the daemon and then receive replies.

The telnet client

By default when you use the command telnet jasper, the telnet client program will attempt to connect to port 23 of the host jasper (23 is the telnet port as listed in /etc/services).

David Jones (20.01.00)

Page 420

85321, Systems Administration

Chapter 16: Network Applications

It is possible to use the telnet client program to connect to other ports. For example the command telnet jasper 25 will connect to port 25 of the machine jasper. The usefulness and problem with this will be discussed on the next couple of pages.

Network protocols
Each network service generally uses its own network protocol that specifies the services it offers, how those services are requested and how they are supplied. For example, the ftp protocol defines the commands that can be used to move files from machine to machine. When you use a command line ftp client, the commands you use are part of the ftp protocol.

Request for comment (RFCs)

For protocols to be useful, both the client and daemon must agree on using the same protocol. If they talk different protocols then no communication can occur. The standards used on the Internet, including those for protocols, are commonly specified in documents called Request for Comments (RFCs). (Not all RFCs are standards). Someone proposing a new Internet standard will write and submit an RFC. The RFC will be distributed to the Internet community who will comment on it and may suggest changes. The standard proposed by the RFC will be adopted as a standard if the community is happy with it. Protocol RFC FTP Telnet SMTP DNS TCP UDP 959 854 821 1035 793 768

Table 16.4 RFCs for Protocols

Table 16.4 lists some of the RFC numbers which describe particular protocols. RFCs can and often are very technical and hard to understand unless you are familiar with the area (the RFC for ftp is about 80 pages long). Exercises
16.4.

Take a look at http://www.faqs.org/ the maintain a collection of FAQs from Usenet news and also provide access to the RFCs. Use this site to view the RFC for SMTP. Take a look through it to get an idea of what is there. The direct URL you want is http://www.faqs.org/rfcs/rfc821.html (at least at the time of writing).

David Jones (20.01.00)

Page 421

85321, Systems Administration

Chapter 16: Network Applications

Text based protocols

Some of these protocols smtp ftp nntp http are text based. They make use of simple text-based commands to perform their duty. Table 16.5 contains a list of the commands that smtp understands. smtp (simple mail transfer protocol) is used to transport mail messages across a TCP/IP network. Command
HELO hostname MAIL FROM: sender-address TO: recipient-address VRFY address EXPN address DATA

Purpose startup and give your hostname mail is coming from this address please send it to this address does this address actually exist (verify) expand this address Im about to start giving you the body of the mail message oops, reset the state and drop the current mail message do nothing set debugging level give me some help please close this connection
Table 16.5 SMTP commands

RSET

NOOP DEBUG [level] HELP QUIT

How it works
When transferring a mail message a client (such as Eudora) will connect to the SMTP daemon (on port 25). The client will then carry out a conversation with the daemon using the commands from Table 16.5. Since these commands are just straight text you can use telnet to simulate the actions of an email client. Doing this actually has some real use. I often use this ability to check on a mail address or to expand a mail alias. The following shows an example of how I might do this. The text in bold is what Ive typed in. The text in italics are comments Ive added after the fact.
beldin:~$ telnet localhost 25 Trying 127.0.0.1... Connected to localhost. Escape character is ^]. 220-beldin.cqu.edu.au Sendmail 8.6.12/8.6.9 ready at Wed, 1 May 1996 13:20:10 +1 000 220 ESMTP spoken here vrfy david check the address david

David Jones (20.01.00)

Page 422

85321, Systems Administration

Chapter 16: Network Applications

250 David Jones <[email protected] vrfy joe check the address joe 550 joe... User unknown vrfy postmaster check the address postmaster 250 <[email protected] expn postmaster postmaster is usually an alias, who is it really?? 250 root <[email protected]

Since 1996, when the above exercise was performed, the Internet has changed a lot. Someof the features shown above may not be supported by some mail servers due to concerns about security and mail spamming (where you are sent email you didnt ask for from people you dont know, usually trying to get you to give them money). Mail spoofing This same approach can be used to spoof mail, that is, send email as someone you are not. This is one of problems with Internet mail. The following is an example of how its done.
bash$ telnet aldur 25 connect to the smtp port (see /etc/services) Trying 138.77.36.29 ... Connected to aldur.cqu.edu.au. Escape character is ^]. 220 aldur.cqu.edu.au Amix Smail3.1.28.1 #2 ready at Sun, 28 Aug 94 12:04 EST helo aldur tell the machine who I am (the name of another machine not a user) 250 aldur.cqu.edu.au Hello aldur mail from: [email protected] this is who the mail is coming from 250 <god@heaven> ... Sender Okay data I want to enter some data which is the message 503 Need RCPT (recipient) cant do that yet, must tell it who to send message to rcpt: david@aldur 500 Command unrecognized oops, typed it wrong rcpt to: david@aldur 250 <david@aldur> ... Recipient Okay data 354 Enter mail, end with "." on a line by itself You have been a naughty boy type in the message . 250 Mail accepted quit bye, bye 221 aldur.cqu.edu.au closing connection Connection closed by foreign host.

There are methods which can be used to identify email sent in this way.

Exercises
16.5.

Using the "telnet" approach connect to an ftp daemon and a http daemon. What commands do they recognise? You might want to refer to the RFCs for those protocols to find out.

David Jones (20.01.00)

Page 423

85321, Systems Administration

Chapter 16: Network Applications

Security
Putting your computer on a network, especially the Internet, makes it accessible to a lot of other people and not all of those people are nice. It is essential that you put in place some sort of security to protect your system from these nasty people. The next chapter takes a more indepth look at security. In this section we examine some of the steps you can take to increase the security of your system.

TCPWrappers/tcpd
The following are entries from two different /etc/inetd.conf files. Both are the entries dealing with the telnet service. The second entry is from a "modern" Linux machine, the first is from an earlier UNIX machine.
telnet stream tcp nowait root /usr/sbin/in.telnetd in.telnetd telnet stream tcp nowait root /usr/sbin/tcpd /usr/sbin/in.telnetd

The difference
Do you notice the difference? The program being run on the Linux machine is
/usr/sbin/tcpd. If you examine the entries in a Linux machines /etc/inetd.conf you will find that this program is executed for all

(almost)

network services. is the public domain program TCPWrappers that comes standard on all Linux machines. It is a special daemon that provides some additional services including added security, access control and logging facilities for all network connections. TCPWrappers works by being inserted between the inetd daemon and the various network daemons that are executed by inetd.
tcpd

Figures 16.1 and 16.2 demonstrate the difference.

Figure

16.1

inetd by itself

David Jones (20.01.00)

Page 424

85321, Systems Administration

Chapter 16: Network Applications

Figure

16.2

inetd with tcpd

tcpd features
tcpd works as follows

a request for a particular network request is received, the configuration of inetd is such that tcpd is executed rather than the actual daemon for this request, tcpd logs the request via syslog, On RedHat each connection is logged into the file /var/log/secure. Information stored includes the time it was made, the host trying to make the connection and the name of the network service being requested. An example entry looks like tcpd then performs a number of checks, These checks make use of some the extra features of tcpd including pattern-based access control. This allows you to specify which hosts are allowed (or not) to use a particular network service. You can use this feature to restrict who can make use of your network services. tcpd also allows you to execute UNIX commands when a particular type of connection occurs.

May

1 12:13:46 beldin in.telnetd[684]: connect from localhost

Exercises
16.6.

The manual page for tcpd says that more information about the access control features of tcpd can be found on the hosts_access(5) manual page. What command would you use to view this page?

hostname verification, Some of the network protocols rely on hostnames for authentication. For example, you may only be able to use the rsh command if your computer is called beldin.cqu.edu.au. It is possible for people to setup computers that will pretend to be another hostname. tcpd offers a feature which will verify that a host is really who they say they are. protection against host address spoofing. It is also possible to spoof an IP address. That is, packets being sent from machine are modified to look as if they are being sent from another,
Page 425

David Jones (20.01.00)

85321, Systems Administration

Chapter 16: Network Applications

trusted, machine. tcpd offers a feature to detect and reject any connections of this type. Exercises
16.7.

Using tcpd how would you achieve the following Configure your machine so there are no network services available. Once you've done this attempt to telnet and ftp to your machine. Keep this tcpd configuration for all the exercises in this group. What effect would the previous question have on the ability for your machine to receive email? Modify your tcpd configuration to allow the receipt of email. Try connecting to the Web daemon on your machine. Assuming you have a standard RedHat installation you should still be able to connect to the Web daemon. Why can you still do this? Shouldn't your tcpd configuration have stopped this?

16.8.

16.9. 16.10.

Other methods for securing a network connection are discussed in the security chapter.

Whats an Intranet?
Intranets are the latest buzzword in the computer industry. The buzzword makers have finally realised the importance of the Internet (and the protocols with which it was constructed) and have started adopting it for a number of purposes. An intranet is basically a local area network used by an organisation that uses the Internet protocols to provide the services normally associated with a LAN plus offering Internet services (but not necessarily Internet access).

Services on an Intranet
The following is a list of the most common services that an Intranet might supply (by no means all of them). This is the list of services we'll discuss in more detail in this chapter. The list includes file sharing, The common ability to share access to applications and data files. It's much simpler to install one copy of an application on a network daemon than it is to install 35 copies on each individual PC. print sharing, and The ability for many different machines to share a printer. It is especially economically if the printer is an expensive, good quality printer. electronic mail. Sometimes called messaging. Electronic mail is fast becoming an essential tool for most businesses. Web serving

David Jones (20.01.00)

Page 426

85321, Systems Administration

Chapter 16: Network Applications

File and print sharing

There is a famous saying in the computing field. The nice thing about standards is that there are so many to choose from. This statement is especially true in the area of sharing printers and files in a local area network. Some of the different protocols are outlined in Table 16.6 which also describes the origins of each protocol. Name Server Message Block (SMB) Netware Description The protocol used by Windows for Workgroups, 95 and NT and OS/2 and a couple of others. Becoming the protocol with the largest number of clients. Netware is the term used to describe Novells network OS. Includes the protocols IPX and NCP (amongst others). A very popular, but possibly dying, network operating system (NOS). The networking built-in to all Macintosh computers. Many Macs now use MacTCP which allows them to "talk" TCP/IP.

Appletalk

Network File System The traditional UNIX based file sharing system. NFS clients (NFS) and daemons are available for most platforms.
Protocols for Table 16.6 sharing files and printers

The "native" form of file sharing on a UNIX machine is NFS. If you wanted to share files between UNIX machines, NFS would be the choice. Due to a number of free software packages, Linux, and most versions of UNIX, can actually act as a server for all of the protocols listed above. Due to the popularity of the Windows family of operating systems, the following will examine the SMB protocols.

Samba
Samba is a piece of software, originally written by Andrew Tridgell (a resident of Canberra), and now maintained by a large number of people from throughout the world. Samba allows a UNIX machine to act as a file and print server for clients running Windows for Workgroups, Windows 95, NT and a couple of other operating systems. The combination of Linux and Samba is possibly the cheapest way of obtaining a server for a Intranet. The following is a very simple introduction to how you might use Samba on a RedHat machine. This process is much simpler on RedHat as Samba comes pre-configured. The readings down below provide much more information about Samba.

David Jones (20.01.00)

Page 427

85321, Systems Administration

Chapter 16: Network Applications

The configuration file for Samba is /etc/smb.conf. An entry in this configuration file which allows a users home directory to be exported to SMB clients is the following
[homes] comment = Home Directories browseable = no read only = no preserve case = yes short preserve case = yes create mode = 0750

If your Linux machine happens to be on a network and you have a Win95/NT or even 3.11 machine on the same network, you should be able to connect to your home directory from that Windows machine using the standard approach for mapping a network drive. Figure 16.3 is the dialog box on a Windows 95 machine.

Dialog

box for

Figure 16.3 mapping a network drive.

In this example, the name of my Linux computer is beldin and my username on beldin is david. Once connected, I can now read and write files from my home directory from within Windows. Chances are most of you will not have a local area network (LAN) at home that has your RedHat Linux machine and another Windows machine connected. This makes it difficult for you to recreate the above example. Luckily Samba comes with a program called smbclient. smbclient is a UNIX program which allows you to connect to Samba shares. This means when you use smbclient you are simulating what would happen if you were using a Windows machine. The following is an example of using smbclient to connect to the same share as in the Windows example above.
[david@beldin david]$ smbclient \\beldin\david Added interface ip=138.77.36.28 bcast=138.77.36.255 nmask=255.255.255.0 Unknown socket option TCP_NODELAY Server time is Fri Feb 6 14:04:50 1998 Timezone is UTC+10.0 Password: Domain=[WORKGROUP] OS=[Unix] Server=[Samba 1.9.17p4] security=user smb: \> help ls dir lcd cd pwd get mget put mput rename more mask del rm mkdir md rmdir rd pq prompt recurse translate lowercase print printmode queue qinfo cancel stat quit q exit newer archive tar blocksize tarmode setmode help ?

David Jones (20.01.00)

Page 428

85321, Systems Administration

Chapter 16: Network Applications

! smb: \> ls *.pdf ei010106.pdf ei020102.pdf ei020103.pdf

129777 229292 291979

Mon Jan 26 12:34:06 1998 Mon Jan 26 12:34:54 1998 Mon Jan 26 12:35:22 1998

50176 blocks of size 16384. 2963 blocks available smb: \>

Once you connect with smbclient you see the smbclient prompt at which you can enter a number of commands. This acts a bit like a commandline ftp prompt. Rather than use the fairly cumbersome smbclient interface to SMB drives Linuxs virtual file system comes to the rescue. With the comment smbmount you can connect SMB drives from Windows machines to your Linux machine and use them as you would any other drive. Exercise
16.11.

Check that Samba is installed and configured on your system. Use smbclient or a Windows machine to see if you can connect to your home directory. This chapter suggests that any network application can be broken down into ports, daemons, clients and protocols. Referring to the above discussion and other available documentation what are the ports, daemons, clients and protocols involved in using Samba on a Linux machine.

16.12.

Email
Electronic mail, at least on the surface, looks fairly easy. However there are a number of issues that make configuring and maintaining Internet electronic mail a complex and occasionally frustrating task. Examining this task in-depth is beyond the scope of this subject. Instead, the following pages will provide an overview of the electronic mail system.

Email components
Programs that help send, reply and distribute email are divided into three categories mail user agents (MUA), These are the programs that people use to read and send email. Common MUAs include Eudora, Netscape (it has a mail and news reader as well as a Web browser) and text-based tools such as elm or pine. MUAs allow a user to read and write email. mail delivery agents (MDA), Once a mail message is delivered to the right computer, the MDA is responsible for placing it into the appropriate mail file. mail transport agents (MTA). Perform a number of tasks including some delivery, forwarding of email to
Page 429

David Jones (20.01.00)

85321, Systems Administration

Chapter 16: Network Applications

other MTAs closer to the final recipient and some address translation. Figure 16.4 provides an overview of how these components fit together.

Figure 16.4 An overview of the mail

system

The following is a brief description of how email is delivered for most people Mail daemon Most people will have an account on a mail server which will be running UNIX, Windows NT or some other operating system. At a minimum, the users account will include a mail file. All email delivered for that user is appended onto the end of that mail file. Remote mail client Reading and writing mail for most people is done using a MUA like Eudora or Netscape on a remote mail client. This "remote mail client" is the users normal computer they use for normal applications. The client mail computer will retrieve the users mail from the mail server using a protocol such as POP or IMAP (see Table 16.6). Sending email will be via the SMTP protocol to the mail servers SMTP daemon (sendmail if its the server is a UNIX computer).

Email Protocols
Table 16.7 lists some of the common protocols associated with email and briefly describes their purpose.
David Jones (20.01.00) Page 430

85321, Systems Administration

Chapter 16: Network Applications

Protocol SMTP POP

Description Simple Mail Transport Protocol, the protocol used to transport mail from one Internet host to another Post Office Protocol, defines a method by which a small host can obtain mail from a larger host without running a MTA (like sendmail). Described in RFCs 1725 1734 Internet Message Access Protocol, allows client mail programs to access and manipulate electronic mail messages on a server, including the manipulation of folders. Described in RFCs 1730, 1731. Multipurpose Internet Mail Extensions, defines methods for sending binary data such as Word documents, pictures and sounds via Internet email which is distributed as text. Described in RFCs 1521 1522 and others. Privacy-Enhanced Mail, message encryption and authentication procedures, proposed standard outlined in RFCs 1421, 1422 and 1423

IMAP

MIME

PEM

Format of text The standard format of Internet email which is described in RFC822 messages
Table 16.7 Protocols and standards associated with Email

Unix mail software Your RedHat Linux machine will include the following software related to email sendmail sendmail is the UNIX MTA. It may well be one of the most difficult and hated pieces of software in the world. However, recent versions have solved many of its problems. sendmail is the SMTP daemon on most UNIX machines. That is it is the server that handles SMTP requests. popd The pop daemon is contacted by MTAs such as Eudora when they wish to transfer a users email from the server onto the client. imapd The imap daemon may not be installed on all machines but it is distributed with RedHat. imapd responds to MTAs which use imap to transfer email from the server to the client. The readings below contain a pointer to a document which describes the differences between IMAP and POP. various mail clients A RedHat machine will include a number of mail clients including mutt, elm, pine, mh, and Netscape.

David Jones (20.01.00)

Page 431

85321, Systems Administration

Chapter 16: Network Applications

Reading The resource materials section on the 85321 Website/CD-ROM has pointers to a number of documents including a sendmail tutorial and a comparison of IMAP and POP. You will need to use these resources for the following exercise.

Exercises
16.13.

Set up email on your Linux machine (refer to the Linux mail HOWTO). Included in the procedure, obtain a POP mail client and get it working. The Netscape web browser includes a POP mail client for UNIX (its what I use to read my mail). The latest versions of Netscape also support IMAP. Configure your system to use IMAP rather than POP.

16.14.

World-Wide Web
The World-Wide Web is the killer application which has really taken the Internet by storm. Most of the Web servers currently on the Internet are UNIX machines running the Apache Web server (http://www.apache.org/). RedHat comes with Apache pre-installed. If you use a Web browser to connect to your Linux machine (e.g. http://localhost/) Redhat provides pointers to documentation on configuring Apache.

Conclusions
This chapter has looked in general at how network services work and in particular at file and print sharing with Samba, email and World-Wide Web. Most network services consist of a port daemon client protocol.

Client programs communicate using an agreed upon protocol with daemons which await requests on a particular port. Network ports are used to deliver information to one of the many network applications that may be running on a computer. Network ports from 0-1024 are used for pre-defined purposes. The allocation of those ports to applications is done in the /etc/services file. The netstat command can be used to examine the currently active network connections including which ports are being used.

David Jones (20.01.00)

Page 432

85321, Systems Administration

Chapter 16: Network Applications

Network daemons are either started in the system start-up scripts (/etc/rc.d/*) or by the inetd daemon. The file /etc/inetd.conf is used to configure which servers inetd will start. Most Linux systems come already installed with tcpd (TCPWrappers). tcpd works with inetd to provide a number of additional features including logging, user validation and access control. Intranets are the latest industry buzzword and are simply a local area network built using Internet protocols. Linux in conjunction with Samba and other public domain tools can act as a very cheap Intranet server offering file and print services, WWW server, electronic mail, ftp and other Internet services. Samba is a public domain piece of software that enables a UNIX computer to act as a file and printer server for client machines running Windows and other LanManager clients.

Review Questions
16.1 Explain the role each of the following play in UNIX networking 1) /etc/services 2) /etc/inetd.conf 3) inetd 4) tcpd 16.2 Youve just obtained the daemon for WWWWW (the fictious replacement for the WWW). The daemon uses the protocol HTTTTTTP, wants to use port 81 and is likely to get many requests. Outline the steps you would have to complete to install the daemon including the files you would have to modify and why

how you would start the daemon (its a program called htttttpd)

16.3 People have been trying to telnet to your machine server.my.domain. List all the things that could be stopping them from logging in.

David Jones (20.01.00)

Page 433

85321, Systems Administration

Chapter 17: Security

Chapter
Security

Local Introduction
The following reading is taken from the Security HOW-TO by Kevin Fenzi and Dave Wreski as part of the Linux Documentation Project. It offers a much better coverage of the material than the original, locally produced chapter (that material is available from the 85321 website/CD-ROM if you feel the need to do a comparison or want an alternative discussion of the data). As you read through the following think about How the advice included here would change the way your personal Linux computer is currently configured? How this advice would change the way you managed a server in a small organisation?

The HOWTO itself mentions a wide range of other resources you can use to get more information about the topic of Security.

Linux Security HOWTO

Kevin Fenzi, [email protected] & Dave Wreski, [email protected] v1.0.2, 25 April 1999 This document is a general overview of security issues that face the administrator of Linux systems. It covers general security philosophy and a number of specific examples of how to better secure your Linux system from intruders. Also included are pointers to security-related material and programs. Improvements, constructive criticism, additions and corrections are gratefully accepted. Please mail your feedback to both authors, with "Security HOWTO" in the subject.

Introduction
This document covers some of the main issues that affect Linux security. General philosophy and net-born resources are discussed. A number of other HOWTO documents overlap with security issues, and those documents have been pointed to wherever appropriate. This document is not meant to be a up to date exploits document. Large numbers of new exploits happen all the time. This document will tell you where to look for such up to date information, and will give some general methods to prevent such exploits from taking place.
David Jones (20.01.00) Page 434

85321, Systems Administration

Chapter 17: Security

New Versions of this Document

New versions of this document will be periodically posted to comp.os.linux.answers. They will also be added to the various anonymous FTP sites that archive such information, including: ftp://metalab.unc.edu/pub/Linux/docs/HOWTO In addition, you should generally be able to find this document on the Linux World Wide Web home page via: http://metalab.unc.edu/mdw/linux.html Finally, the very latest version of this document should also be available in various formats from: http://scrye.com/~kevin/lsh/

Feedback
All comments, error reports, additional information and criticism of all sorts should be directed to: [email protected] and [email protected] Note: Please send your feedback to both authors. Also, be sure and include "Linux" "security", or "HOWTO" in your subject to avoid Kevins spam filter.

Disclaimer
No liability for the contents of this document can be accepted. Use the concepts, examples and other content at your own risk. Additionally, this is an early version, possibly with many inaccuracies or errors. A number of the examples and descriptions use the RedHat(tm) package layout and system setup. Your mileage may vary. As far as we know, only programs that, under certain terms may be used or evaluated for personal purposes will be described. Most of the programs will be available, complete with source, under GNU <http://www.gnu.org/copyleft/gpl.html> terms.

Copyright Information
This document is copyrighted (c)1998,1999 Kevin Fenzi and Dave Wreski, and distributed under the following terms: Linux HOWTO documents may be reproduced and distributed in whole or in part, in any medium, physical or electronic, as long as this copyright notice is retained on all copies. Commercial redistribution is allowed and encouraged; however, the authors would like to be notified of any such distributions. All translations, derivative works, or aggregate works incorporating any Linux HOWTO documents must be covered under this copyright notice. That is, you may not produce a derivative work from a HOWTO and
Page 435

David Jones (20.01.00)

85321, Systems Administration

Chapter 17: Security

impose additional restrictions on its distribution. Exceptions to these rules may be granted under certain conditions; please contact the Linux HOWTO coordinator at the address given below. If you have questions, please contact Tim Bynum, the Linux HOWTO coordinator, at [email protected]

Overview
This document will attempt to explain some procedures and commonly-used software to help your Linux system be more secure. It is important to discuss some of the basic concepts first, and create a security foundation, before we get started.

Why Do We Need Security?

In the ever-changing world of global data communications, inexpensive Internet connections, and fast-paced software development, security is becoming more and more of an issue. Security is now a basic requirement because global computing is inherently insecure. As your data goes from point A to point B on the Internet, for example, it may pass through several other points along the way, giving other users the opportunity to intercept, and even alter, it. Even other users on your system may maliciously transform your data into something you did not intend. Unauthorized access to your system may be obtained by intruders, also known as "crackers", who then use advanced knowledge to impersonate you, steal information from you, or even deny you access to your own resources. If youre wondering what the difference is between a "Hacker" and a "Cracker", see Eric Raymonds document, "How to Become A Hacker", available at http://sagan.earthspace.net/~esr/faqs/hackerhowto.html.

How Secure Is Secure?

First, keep in mind that no computer system can ever be "completely secure". All you can do is make it increasingly difficult for someone to compromise your system. For the average home Linux user, not much is required to keep the casual cracker at bay. For high profile Linux users (banks, telecommunications companies, etc), much more work is required. Another factor to take into account is that the more secure your system is, the more intrusive your security becomes. You need to decide where in this balancing act your system will still usable, and yet secure for your purposes. For instance, you could require everyone dialing into your system to use a callback modem to call them back at their home number. This is more secure, but if someone is not at home, it makes it difficult for them to login. You could also setup your Linux system with no network or connection to the Internet, but this limits its usefulness. If you are a large to medium-sized site, you should establish a security policy stating how much security is required by your site and what auditing is in place to check it. You can find a well-known security policy example at http://ds.internic.net/rfc/rfc2196.txt. It has been recently updated, and contains a great framework for establishing a security policy for your company.
David Jones (20.01.00) Page 436

85321, Systems Administration

Chapter 17: Security

What Are You Trying to Protect?

Before you attempt to secure your system, you should determine what level of threat you have to protect against, what risks you should or should not take, and how vulnerable your system is as a result. You should analyze your system to know what youre protecting, why youre protecting it, what value it has, and who has responsibility for your data and other assets. Risk is the possibility that an intruder may be successful in attempting to access your computer. Can an intruder read or write files, or execute programs that could cause damage? Can they delete critical data? Can they prevent you or your company from getting important work done? Dont forget: someone gaining access to your account, or your system, can also impersonate you. Additionally, having one insecure account on your system can result in your entire network being compromised. If you allow a single user to login using a .rhosts file, or to use an insecure service, such as tftp, you risk an intruder getting his foot in the door. Once the intruder has a user account on your system, or someone elses system, it can be used to gain access to another system, or another account. Threat is typically from someone with motivation to gain unauthorized access to your network or computer. You must decide who you trust to have access to your system, and what threat they could pose. There are several types of intruders, and it is useful to keep their different characteristics in mind as you are securing your systems. The Curious - This type of intruder is basically interested in finding out what type of system and data you have.

The Malicious - This type of intruder is out to either bring down your systems, or deface your web page, or otherwise force you to spend time and money recovering from the damage he has caused. The High-Profile Intruder - This type of intruder is trying to use your system to gain popularity and infamy. He might use your highprofile system to advertise his abilities. The Competition - This type of intruder is interested in what data you have on your system. It might be someone who thinks you have something that could benefit him, financially or otherwise. The Borrowers - This type of intruder is interested in setting up shop on your system and using its resources for their own purposes. They typically will run chat or irc servers, porn archive sites, or even DNS servers. The Leapfrogger - This type of intruder is only interested in your system to use it to get into other systems. If your system is well connected or a gateway to a number of internal hosts, you may well see this type trying to compromise your system.

Vulnerability describes how well-protected your computer is from another network, and the potential for someone to gain unauthorized access. Whats at stake if someone breaks into your system? Of course the concerns of a dynamic PPP home user will be different from those of a company connecting
David Jones (20.01.00) Page 437

85321, Systems Administration

Chapter 17: Security

their machine to the Internet, or another large network. How much time would it take to retrieve/recreate any data that was lost? An initial time investment now can save ten times more time later if you have to recreate data that was lost. Have you checked your backup strategy, and verified your data lately?

Developing A Security Policy

Create a simple, generic policy for your system that your users can readily understand and follow. It should protect the data youre safeguarding as well as the privacy of the users. Some things to consider adding are: who has access to the system (Can my friend use my account?), whos allowed to install software on the system, who owns what data, disaster recovery, and appropriate use of the system. A generally accepted security policy starts with the phrase That which is not permitted is prohibited This means that unless you grant access to a service for a user, that user shouldnt be using that service until you do grant access. Make sure the policies work on your regular user account. Saying, "Ah, I cant figure this permissions problem out, Ill just do it as root" can lead to security holes that are very obvious, and even ones that havent been exploited yet. rfc1244 is a document that describes how to create your own network security polity. rfc1281 is a document that shows an example security policy with detailed descriptions of each step. Finally, you might want to look at the COAST policy archive at ftp://coast.cs.purdue.edu/pub/doc/policy to see what some real life security policies look like.

Means of Securing Your Site

This document will discuss various means with which you can secure the assets you have worked hard for: your local machine, your data, your users, your network, even your reputation. What would happen to your reputation if an intruder deleted some of your users data? Or defaced your web site? Or published your companys corporate project plan for next quarter? If you are planning a network installation, there are many factors you must take into account before adding a single machine to your network. Even if you have a single dialup PPP account, or just a small site, this does not mean intruders wont be interested in your systems. Large, high profile sites are not the only targets -- many intruders simply want to exploit as many sites as possible, regardless of their size. Additionally, they may use a security hole in your site to gain access to other sites youre connected to. Intruders have a lot of time on their hands, and can avoid guessing how youve obscured your system just by trying all the possibilities. There are also a number of reasons an intruder may be interested in your systems, which we will discuss later.

David Jones (20.01.00)

Page 438

85321, Systems Administration

Chapter 17: Security

Host Security Perhaps the area of security on which administrators concentrate most is hostbased security. This typically involves making sure your own system is secure, and hoping everyone else on your network does the same. Choosing good passwords, securing your hosts local network services, keeping good accounting records, and upgrading programs with known security exploits are among the things the local security administrator is responsible for doing. Although this is absolutely necessary, it can become a daunting task once your network becomes larger than a few machines. Network Security Network security is also as necessary as local host security. With hundreds, thousands, or more computers on the same network, you cant rely on each one of those systems being secure. Ensuring that only authorized users can use your network, building firewalls, using strong encryption, and ensuring there are no "rogue" (that is, unsecured) machines on your network are all part of the network security administrators duties. This document will discuss some of the techniques used to secure your site, and hopefully show you some of the ways to prevent an intruder from gaining access to what you are trying to protect. Security Through Obscurity One type of security that must be discussed is "security through obscurity". This means, for example, moving a service that has known security vunerabilities to a non standard port in hopes that attackers wont notice its there and thus wont exploit it. Rest assured that they can determine that its there and will exploit it. Security through obscurity is no security at all. Simply because you may have a small site, or a relatively low profile, does not mean an intruder wont be interested in what you have. Well discuss what youre protecting in the next sections.

Organization of This Document

This document has been divided into a number of sections. They cover several broad security issues. The first, Physical Security, covers how you need to protect your physical machine from tampering. The second, Local Security, describes how to protect your system from tampering by local users. The third, Files and Filesystem Security, shows you how to setup your filesystems and premissions on your files. The next, Password Security and Encryption, discusses how to use encryption to better secure your machine and network. Kernel Security discusses what kernel options you should set or be aware of for a more secure system. Network Security, describes how to better secure your Linux system from network attacks. Security Preparation, discusses how to prepare your machine(s) before bringing them on-line. Next, What To Do During and After a Break-in, discusses what to do when you detect a system compromise in progress or detect one that has recently happened. In Security Resources, some primary security resources are enumerated. The Q and A section Frequently Asked Questions, answers some frequently asked questions, and finally a conclusion in Conclusion section.
David Jones (20.01.00) Page 439

85321, Systems Administration

Chapter 17: Security

The two main points to realize when reading this document are: Be aware of your system. Check system logs such as /var/log/messages and keep an eye on your system, and Keep your system up to date by making sure you have installed the current versions of software and have upgraded per security alerts. Just doing this will help make your system markedly more secure.

Physical Security
The first layer of security you need to take into account is the physical security of your computer systems. Who has direct physical access to your machine? Should they? Can you protect your machine from their tampering? Should you? How much physical security you need on your system is very dependent on your situation, and/or budget. If you are a home user, you probably dont need a lot (although you might need to protect your machine from tampering by children or annoying relatives). If you are in a Lab, you need considerably more, but users will still need to be able to get work done on the machines. Many of the following sections will help out. If you are in an office, you may or may not need to secure your machine off hours or while you are away. At some companies, leaving your console unsecured is a termination offense. Obvious physical security methods such as locks on doors, cables, locked cabinets, and video surveillance are all good ideas, but beyond the scope of this document. :)

Computer locks
Many modern PC cases include a "locking" feature. Usually this will be a socket on the front of the case that allows you to turn an included key to a locked or unlocked position. Case locks can help prevent someone from stealing your PC, or opening up the case and directly manipulating/stealing your hardware. They can also sometimes prevent someone from rebooting your computer on their own floppy or other hardware. These case locks do different things according to the support in the motherboard and how the case is constructed. On many PCs they make it so you have to break the case to get the case open. On some others, they make it so that it will not let you plug in new keyboards and mice. Check your motherboard or case instructions for more information. This can sometimes be a very useful feature, even though the locks are usually very low quality and can easily be defeated by attackers with locksmithing. Some cases (most notably SPARCs and macs) have a dongle on the back that, if you put a cable through attackers would have to cut the cable or break the case to get into it. Just putting a padlock or combo lock through these can be a good deterrent to someone stealing your machine.

David Jones (20.01.00)

Page 440

85321, Systems Administration

Chapter 17: Security

BIOS Security
The BIOS is the lowest level of software that configures or manipulates your x86-based hardware. LILO and other Linux boot methods access the BIOS to determine how to boot up your Linux machine. Other hardware that Linux runs on has similar software (OpenFirmware on Macs and new Suns, Sun boot PROM, etc...). You can use your BIOS to prevent attackers from rebooting your machine and manipulating your Linux system. Many PC BIOSs let you set a boot password. This doesnt provide all that much security (the BIOS can be reset, or removed if someone can get into the case), but might be a good deterrent (i.e. it will take time and leave traces of tampering). Similarly, on S/Linux (Linux for SPARC(tm) porcessor machines), your EEPROM can be set to require a boot-up password. This might slow attackers down. Many x86 BIOSs also allow you to specify various other good security settings. Check your BIOS manual or look at it the next time you boot up. For example, some BIOSs disallow booting from floppy drives and some require passwords to access some BIOS features. Note: If you have a server machine, and you set up a boot password, your machine will not boot up unattended. Keep in mind that you will need to come in and supply the password in the event of a power failure. ;(

Boot Loader Security

The various Linux boot loaders also can have a boot password set. LILO, for example, has password and restricted settings; password always requires password at boot time, whereas restricted requires a boot-time password only if you specify options (such as single) at the LILO prompt. Keep in mind when setting all these passwords that you need to remember them. :) Also remember that these passwords will merely slow the determined attacker. They wont prevent someone from booting from a floppy, and mounting your root partition. If you are using security in conjunction with a boot loader, you might as well disable booting from a floppy in your computers BIOS, and password-protect the BIOS. If anyone has security-related information from a different boot loader, we would love to hear it. (grub, silo, milo, linload, etc). Note: If you have a server machine, and you set up a boot password, your machine will not boot up unattended. Keep in mind that you will need to come in and supply the password in the event of a power failure. ;( xlock and vlock If you wander away from your machine from time to time, it is nice to be able to "lock" your console so that no one tampers with or looks at your work. Two programs that do this are: xlock and vlock. xlock is a X display locker. It should be included in any Linux distributions that support X. Check out the man page for it for more options, but in general you can run xlock from any xterm on your console and it will lock the display and require your password to unlock.
David Jones (20.01.00) Page 441

85321, Systems Administration

Chapter 17: Security

vlock is a simple little program that allows you to lock some or all of the virtual consoles on your Linux box. You can lock just the one you are working in or all of them. If you just lock one, others can come in and use the console; they will just not be able to use your virtual console until you unlock it. vlock ships with redhat Linux, but your mileage may vary. Of course locking your console will prevent someone from tampering with your work, but wont prevent them from rebooting your machine or otherwise disrupting your work. It also does not prevent them from accessing your machine from another machine on the network and causing problems. More importantly, it does not prevent someone from switching out of the X Window System entirely, and going to a normal virtual console login prompt, or to the VC that X11 was started from, and suspending it, thus obtaining your priviledges. For this reason, you might consider only using it while under control of xdm.

Detecting Physical Security Compromises

The first thing to always note is when your machine was rebooted. Since Linux is a robust and stable OS, the only times your machine should reboot is when you take it down for OS upgrades, hardware swapping, or the like. If your machine has rebooted without you doing it, that may be a sign that an intruder has compromised it. Many of the ways that your machine can be compromised require the intruder to reboot or power off your machine. Check for signs of tampering on the case and computer area. Although many intruders clean traces of their presence out of logs, its a good idea to check through them all and note any discrepancy. It is also a good idea to store log data at a secure location, such as a dedicated log server within your well-protected network. Once a machine has been compromised, log data becomes of little use as it most likely has also been modified by the intruder. The syslog daemon can be configured to automatically send log data to a central syslog server, but this is typically sent in cleartext data, allowing an intruder to view data as it is being transferred. This may reveal information about your network that is not intended to be public. There are syslog daemons available that encrypt the data as it is being sent. Also be aware that faking syslog messages is easy - with an exploit program having been published. Syslog even accepts net log entries claiming to come from the local host without indicating their true origin. Some things to check for in your logs: Short or incomplete logs. Logs containing strange timestamps. Logs with incorrect permissions or ownership. Records of reboots or restarting of services. missing logs. su entries or logins from strange places.

We will discuss system log data later in the HOWTO.

David Jones (20.01.00) Page 442

85321, Systems Administration

Chapter 17: Security

Local Security
The next thing to take a look at is the security in your system against attacks from local users. Did we just say local users? Yes! Getting access to a local user account is one of the first things that system intruders attempt while on their way to exploiting the root account. With lax local security, they can then "upgrade" their normal user access to root access using a variety of bugs and poorly setup local services. If you make sure your local security is tight, then the intruder will have another hurdle to jump. Local users can also cause a lot of havoc with your system even (especially) if they really are who they say they are. Providing accounts to people you dont know or have no contact information for is a very bad idea.

Creating New Accounts

You should make sure to provide user accounts with only the minimal requirements for the task they need to do. If you provide your son (age 10) with an account, you might want him to only have access to a word processor or drawing program, but be unable to delete data that is not his. Several good rules of thumb when allowing other people legitimate access to your Linux machine: Give them the minimal amount of privileges they need. Be aware when/where they login from, or should be logging in from. Make sure to remove inactive accounts The use of the same user-ID on all computers and networks is advisable to ease account maintence, as well as permit easier analysis of log data. The creation of group user-IDs should be absolutely prohibited. User accounts also provide accountability, and this is not possible with group accounts.

Many local user accounts that are used in security compromises are ones that have not been used in months or years. Since no one is using them they, provide the ideal attack vehicle.

Root Security
The most sought-after account on your machine is the root (superuser) account. This account has authority over the entire machine, which may also include authority over other machines on the network. Remember that you should only use the root account for very short, specific tasks, and should mostly run as a normal user. Even small mistakes made while logged in as the root user can cause problems. The less time you are on with root privledges, the safer you will be. Several tricks to avoid messing up your own box as root: When doing some complex command, try running it first in a nondestructive way...especially commands that use globbing: e.g., if you want to do "rm foo*.bak", first do "ls foo*.bak" and make sure you are going to
Page 443

David Jones (20.01.00)

85321, Systems Administration

Chapter 17: Security

delete the files you think you are. Using echo in place of destructive commands also sometimes works. Provide your users with a default alias to the rm command to ask for confirmation for deletion of files. Only become root to do single specific tasks. If you find yourself trying to figure out how to do something, go back to a normal user shell until you are sure what needs to be done by root. The command path for the root user is very important. The command path (that is, the PATH environment variable) specifies the directories in which the shell searches for programs. Try to limit the command path for the root user as much as possible, and never include . (which means "the current directory") in your PATH. Additionally, never have writable directories in your search path, as this can allow attackers to modify or place new binaries in your search path, allowing them to run as root the next time you run that command. Never use the rlogin/rsh/rexec suite of tools (called the r- utilities) as root. They are subject to many sorts of attacks, and are downright dangerous run as root. Never create a .rhosts file for root. The /etc/securetty file contains a list of terminals that root can login from. By default (on Red Hat Linux) this is set to only the local virtual consoles(vtys). Be very careful of adding anything else to this file. You should be able to login remotely as your regular user account and then su if you need to (hopefully over ssh or other encrypted channel), so there is no need to be able to login directly as root. Always be slow and deliberate running as root. Your actions could affect a lot of things. Think before you type!

If you absolutely positively need to allow someone (hopefully very trusted) to have root access to your machine, there are a few tools that can help. sudo allows users to use their password to access a limited set of commands as root. This would allow you to, for instance, let a user be able to eject and mount removable media on your Linux box, but have no other root privileges. sudo also keeps a log of all successful and unsuccessful sudo attempts, allowing you to track down who used what command to do what. For this reason sudo works well even in places where a number of people have root access, because it helps you keep track of changes made. Although sudo can be used to give specific users specific privileges for specific tasks, it does have several shortcomings. It should be used only for a limited set of tasks, like restarting a server, or adding new users. Any program that offers a shell escape will give root access to a user invoking it via sudo. This includes most editors, for example. Also, a program as innocuous as /bin/cat can be used to overwrite files, which could allow root to be exploited. Consider sudo as a means for accountability, and dont expect it to replace the root user and still be secure.

David Jones (20.01.00)

Page 444

85321, Systems Administration

Chapter 17: Security

Files and Filesystem Security

A few minutes of preparation and planning ahead before putting your systems online can help to protect them and the data stored on them. There should never be a reason for users home directories to allow SUID/SGID programs to be run from there. Use the nosuid option in /etc/fstab for partitions that are writable by others than root. You may also wish to use nodev and noexec on users home partitions, as well as /var, thus prohibiting execution of programs, and creation of character or block devices, which should never be necessary anyway. If you are exporting filesystems using NFS, be sure to configure /etc/exports with the most restrictive access possible. This means not using wildcards, not allowing root write access, and exporting read-only wherever possible. Configure your users file-creation umask to be as restrictive as possible. See umask settings. If you are mounting filesystems using a network filesystem such as NFS, be sure to configure /etc/exports with suitable restrictions. Typically, using nodev, nosuid, and perhaps noexec, are desireable. Set filesystem limits instead of allowing unlimited as is the default. You can control the per-user limits using the resource-limits PAM module and /etc/pam.d/limits.conf. For example, limits for group users might look like this:
@users @users @users hard hard hard core nproc rss 0 50 5000

This says to prohibit the creation of core files, restrict the number of processes to 50, and restrict memory usage per user to 5M. The /var/log/wtmp and /var/run/utmp files contain the login records for all users on your system. Their integrity must be maintained because it can be used to determine when and from where a user (or potential intruder) has entered your system. These files should also have 644 permissions, without affecting normal system operation. The immutable bit can be used to prevent accidentally deleting or overwriting a file that must be protected. It also prevents someone from creating a symbolic link to the file (such symbolic links have been the source of attacks involving deleting /etc/passwd or /etc/shadow). See the chattr(1) man page for information on the immutable bit. SUID and SGID files on your system are a potential security risk, and should be monitored closely. Because these programs grant special privileges to the user who is executing them, it is necessary to ensure that insecure programs are not installed. A favorite trick of crackers is to exploit SUID-root programs, then leave a SUID program as a backdoor to get in the next time, even if the original hole is plugged. Find all SUID/SGID programs on your system, and keep track of what they are, so you are aware of any changes which could indicate a potential intruder. Use the following command to find all SUID/SGID programs on your system:
Page 445

David Jones (20.01.00)

85321, Systems Administration

Chapter 17: Security

root#

find / -type f \( -perm -04000 -o -perm -02000 \)

The Debian distribution runs a job each night to determine what SUID files exist. It then compairs this to the previous nights run. You can look in /var/log/suid* for this log. You can remove the SUID or SGID permissions on a suspicious program with chmod, then change it back if you absolutely feel it is necessary. World-writable files, particularly system files, can be a security hole if a cracker gains access to your system and modifies them. Additionally, world-writable directories are dangerous, since they allow a cracker to add or delete files as he wishes. To locate all world-writable files on your system, use the following command:
root# find / -perm -2 ! -type l ls

and be sure you know why those files are writable. In the normal course of operation, several files will be world-writable, including some from /dev, and symbolic links, thus the ! -type l which excludes these from the previous find command.

Unowned files may also be an indication an intruder has accessed your system. You can locate files on your system that have no owner, or belong to no group with the command:
root# find / -nouser -o -nogroup -print

Finding .rhosts files should be a part of your regular system administration duties, as these files should not be permitted on your system. Remember, a cracker only needs one insecure account to potentially gain access to your entire network. You can locate all .rhosts files on your system with the following command:
root# find /home -name .rhosts -print

Finally, before changing permissions on any system files, make sure you understand what you are doing. Never change permissions on a file because it seems like the easy way to get things working. Always determine why the file has that permission before changing it.

Umask Settings
The umask command can be used to determine the default file creation mode on your system. It is the octal complement of the desired file mode. If files are created without any regard to their permissions settings, the user could inadvertently give read or write permission to someone that should not have this permission. Typically umask settings include 022, 027, and 077 (which is the most restrictive). Normally the umask is set in /etc/profile, so it applies to all users on the system. The file creation mask can be calculated by subtracting the desired value from 777. In other words, a umask of 777 would cause newly-created files to contain no read, write or execute permission for anyone. A mask of 666 would cause newly-created files to have a mask of 111. For example, you may have a line that looks like this:
# Set the users default umask umask 033
David Jones (20.01.00) Page 446

85321, Systems Administration

Chapter 17: Security

Be sure to make roots umask 077, which will disable read, write, and execute permission for other users, unless explicitly changed using chmod. In this case, newly-created directories would have 744 permissions, obtained by subtracting 033 from 777. Newly-created files using the 033 umask would have permissions of 644. If you are using Red Hat, and adhere to their user and group ID creation scheme (User Private Groups), it is only necessary to use 002 for a umask. This is due to the fact that the default configuration is one user per group.

File Permissions
Its important to ensure that your system files are not open for casual editing by users and groups who shouldnt be doing such system maintenance. Unix seperates access control on files and directories according to three characteristics: owner, group, and other. There is always exactly one owner, any number of members of the group, and everyone else. A quick explanation of Unix permissions: Ownership - Which user(s) and group(s) retain(s) control of the permission settings of the node and parent of the node Permissions - Bits capable of being set or reset to allow certain types of access to it. Permissions for directories may have a different meaning than the same set of permissions on files. Read: To be able to view contents of a file To be able to read a directory

Write: To be able to add to or change a file To be able to delete or move files in a directory

Execute: To be able to run a binary program or shell script To be able to search in a directory, combined with read permission Save Text Attribute: (For directories) The "sticky bit" also has a different meaning when applied to directories than when applied to files. If the sticky bit is set on a directory, then a user may only delete files that the he owns or for which he has explicit write permission granted, even when he has write access to the directory. This is designed for directories like /tmp, which are world-writable, but where it may not be desirable to allow any user to delete files at will. The sticky bit is seen as a t in a long directory listing.

David Jones (20.01.00)

Page 447

85321, Systems Administration

Chapter 17: Security

SUID Attribute: (For Files) This describes set-user-id permissions on the file. When the set user ID access mode is set in the owner permissions, and the file is executable, processes which run it are granted access to system resources based on user who owns the file, as opposed to the user who created the process. This is the cause of many "buffer overflow" exploits. SGID Attribute: (For Files) If set in the group permissions, this bit controls the "set group id" status of a file. This behaves the same way as SUID, except the group is affected instead. The file must be executable for this to have any effect. SGID Attribute: (For directories) If you set the SGID bit on a directory (with chmod g+s directory), files created in that directory will have their group set to the directorys group. You - The owner of the file Group - The group you belong to Everyone - Anyone on the system that is not the owner or a member of the group File Example:
-rw-r--r-- 1 kevin users 114 Aug 28 1997 .zlogin 1st bit - directory? (no) 2nd bit - read by owner? (yes, by kevin) 3rd bit - write by owner? (yes, by kevin) 4th bit - execute by owner? (no) 5th bit - read by group? (yes, by users) 6th bit - write by group? (no) 7th bit - execute by group? (no) 8th bit - read by everyone? (yes, by everyone) 9th bit - write by everyone? (no) 10th bit - execute by everyone? (no)

The following lines are examples of the minimum sets of permissions that are required to perform the access described. You may want to give more permission than whats listed here, but this should describe what these minimum permissions on files do:
-r---------w------Allow read access to the file by owner Allows the owner to modify or delete the file (Note that anyone with write permission to the directory the file is in can overwrite it and thus delete it) The owner can execute this program, but not shell scripts, which still need read permission Will execute with effective User ID = to owner Will execute with effective Group ID = to group No update of "last modified time". Usually used for swap files No effect. (formerly sticky bit)

---x--------s-------------s-rw------T ---t------

David Jones (20.01.00)

Page 448

85321, Systems Administration

Chapter 17: Security

Directory Example:
drwxr-xr-x 3 kevin users 512 Sep 19 13:47 .public_html/ 1st bit - directory? (yes, it contains many files) 2nd bit - read by owner? (yes, by kevin) 3rd bit - write by owner? (yes, by kevin) 4th bit - execute by owner? (yes, by kevin) 5th bit - read by group? (yes, by users 6th bit - write by group? (no) 7th bit - execute by group? (yes, by users) 8th bit - read by everyone? (yes, by everyone) 9th bit - write by everyone? (no) 10th bit - execute by everyone? (yes, by everyone)

The following lines are examples of the minimum sets of permissions that are required to perform the access described. You may want to give more permission than whats listed, but this should describe what these minimum permissions on directories do:
dr-------d--x-----dr-x-----d-wx-----d------x-t d---s--s-The contents can be listed, but file attributes cant be read The directory can be entered, and used in full execution paths File attributes can be read by owner Files can be created/deleted, even if the directory isnt the current one Prevents files from deletion by others with write access. Used on /tmp No effect

System configuration files (usually in /etc) are usually mode 640 (-rw-r-----), and owned by root. Depending on your sites security requirements, you might adjust this. Never leave any system files writable by a group or everyone. Some configuration files, including /etc/shadow, should only be readable by root, and directories in /etc should at least not be accessible by others. SUID Shell Scripts SUID shell scripts are a serious security risk, and for this reason the kernel will not honor them. Regardless of how secure you think the shell script is, it can be exploited to give the cracker a root shell.

Integrity Checking with Tripwire Tripwire

Another very good way to detect local (and also network) attacks on your system is to run an integrity checker like Tripwire. Tripwire runs a number of checksums on all your important binaries and config files and compares them against a database of former, known-good values as a reference. Thus, any changes in the files will be flagged. Its a good idea to install Tripwire onto a floppy, and then physically set the write protect on the floppy. This way intruders cant tamper with Tripwire itself or change the database. Once you have Tripwire setup, its a good idea to run it as part of your normal security administration duties to see if anything has changed. You can even add a crontab entry to run Tripwire from your floppy every night and mail you the results in the morning. Something like:
# set mailto MAILTO=kevin
David Jones (20.01.00) Page 449

85321, Systems Administration

Chapter 17: Security

# run Tripwire 15 05 * * * root /usr/local/adm/tcheck/tripwire

will mail you a report each morning at 5:15am. Tripwire can be a godsend to detecting intruders before you would otherwise notice them. Since a lot of files change on the average system, you have to be careful what is cracker activity and what is your own doing. You can find Tripwire at http://www.tripwiresecurity.com, free of charge. Manuals and support can be purchased.

5.4. Trojan Horses

"Trojan Horses" are named after the fabled ploy in Homers "Iliad". The idea is that a cracker distributes a program or binary that sounds great, and encourages other people to download it and run it as root. Then the program can compromise their system while they are not paying attention. While they think the binary they just pulled down does one thing (and it might very well), it also compromises their security. You should take care of what programs you install on your machine. Redhat provides MD5 checksums and PGP signatures on its RPM files so you can verify you are installing the real thing. Other distributions have similar methods. You should never run any unfamiliar binary, for which you dont have the source, as root! Few attackers are willing to release source code to public scrutiny. Although it can be complex, make sure you are getting the source for a program from its real distribution site. If the program is going to run as root, make sure either you or someone you trust has looked over the source and verified it.

Password Security and Encryption

One of the most important security features used today are passwords. It is important for both you and all your users to have secure, unguessable passwords. Most of the more recent Linux distributions include passwd programs that do not allow you to set a easily guessable password. Make sure your passwd program is up to date and has these features. In-depth discussion of encryption is beyond the scope of this document, but an introduction is in order. Encryption is very useful, possibly even necessary in this day and age. There are all sorts of methods of encrypting data, each with its own set of characteristics. Most Unicies (and Linux is no exception) primarily use a one-way encryption algorithm, called DES (Data Encryption Standard) to encrypt your passwords. This encrypted password is then stored in (typically) /etc/passwd (or less commonly) /etc/shadow. When you attempt to login, the password you type in is encrypted again and compared with the entry in the file that stores your passwords. If they match, it must be the same password, and you are allowed access. Although DES is a two-way encryption algorithm (you can code and then decode a message, given the right keys), the variant that most unices use is one-way. This means that it should not be possible to reverse the encryption to get the password from the contents of /etc/passwd (or /etc/shadow).
David Jones (20.01.00) Page 450

85321, Systems Administration

Chapter 17: Security

Brute force attacks, such as "Crack" or "John the Ripper" (see the later section covering these) can often guess passwords unless your password is sufficiently random. PAM modules (see below) allow you to use a different encryption routine with your passwords (MD5 or the like). You can use Crack to your advantage, as well. Consider periodically running Crack against your own password database, to find insecure passwords. Then contact the offending user, and instruct him to change his password. You can go to http://consult.cern.ch/writeup/security/security_3.html for information on how to choose a good password.

PGP and Public-Key Cryptography

Public-key cryptography, such as that used for PGP, uses one key for encryption, and one key for decryption. Traditional cryptography, however, uses the same key for encryption and decryption; this key must be known to both parties, and thus somehow transferred from one to the other securely. To alleviate the need to securely transmit the encryption key, public-key encryption uses two separate keys: a public key and a private key. Each persons public key is available by anyone to do the encryption, while at the same time each person keeps his or her private key to decrypt messages encrypted with the correct public key. There are advantages to both public key and private key cryptography, and you can read about those differences in the RSA Cryptography FAQ <http://www.rsa.com/rsalabs/newfaq/>, listed at the end of this section. PGP (Pretty Good Privacy) is well-supported on Linux. Versions 2.6.2 and 5.0 are known to work well. For a good primer on PGP and how to use it, take a look at the PGP FAQ: http://www.pgp.com/service/export/faq/55faq.cgi Be sure to use the version that is applicable to your country. Due to export restrictions by the US Government, strong-encryption is prohibited from being transferred in electronic form outside the country. US export controls are now managed by EAR (Export Administration Regulations). They are no longer governed by ITAR. There is also a step-bystep guide for configuring PGP on Linux available at http://mercury.chem.pitt.edu/~angel/LinuxFocus/English/November1997/articl e7.html. It was written for the international version of PGP, but is easily adaptable to the United States version. You may also need a patch for some of the latest versions of Linux; the patch is available at ftp://metalab.unc.edu/pub/Linux/apps/crypto. There is a project working on a free re-implementation of pgp with open source. GnuPG is a complete and free replacement for PGP. Because it does not use IDEA or RSA it can be used without any restrictions. GnuPG is nearly in compliance with RFC2440 (OpenPGP). See the GNU Privacy Guard web page for more information: http://www.gpg.org/. More information on cryptography can be found in the RSA cryptography FAQ, available at http://www.rsa.com/rsalabs/newfaq/. Here you will find information on such terms as "Diffie-Hellman", "public-key cryptography", "digital certificates", etc.
David Jones (20.01.00) Page 451

85321, Systems Administration

Chapter 17: Security

SSL, S-HTTP, HTTPS and S/MIME

Often users ask about the differences between the various security and encryption protocols, and how to use them. While this isnt an encryption document, it is a good idea to explain briefly what each protocol is, and where to find more information. SSL: - SSL, or Secure Sockets Layer, is an encryption method developed by Netscape to provide security over the Internet. It supports several different encryption protocols, and provides client and server authentication. SSL operates at the transport layer, creates a secure encrypted channel of data, and thus can seamlessly encrypt data of many types. This is most commonly seen when going to a secure site to view a secure online document with Communicator, and serves as the basis for secure communications with Communicator, as well as many other Netscape Communications data encryption. More information can be found at http://www.consensus.com/security/ssl-talk-faq.html. Information on Netscapes other security implementations, and a good starting point for these protocols is available at http://home.netscape.com/info/securitydoc.html. S-HTTP: - S-HTTP is another protocol that provides security services across the Internet. It was designed to provide confidentiality, authentication, integrity, and non-repudiability [cannot be mistaken for someone else] while supporting multiple key-management mechanisms and cryptographic algorithms via option negotiation between the parties involved in each transaction. S-HTTP is limited to the specific software that is implementing it, and encrypts each message individually. [ From RSA Cryptography FAQ, page 138] S/MIME: - S/MIME, or Secure Multipurpose Internet Mail Extension, is an encryption standard used to encrypt electronic mail and other types of messages on the Internet. It is an open standard developed by RSA, so it is likely we will see it on Linux one day soon. More information on S/MIME can be found at http://home.netscape.com/assist/security/smime/overview.html.

Linux IPSEC Implementations

Along with CIPE, and other forms of data encryption, there is also several other implementations of IPSEC for Linux. IPSEC is an effort by the IETF to create cryptographically-secure communications at the IP network level, and to provide authentication, integrity, access control, and confidentiality. Information on IPSEC and Internet draft can be found at http://www.ietf.org/html.charters/ipsec-charter.html. You can also find links to other protocols involving key management, and an IPSEC mailing list and archives. The x-kernel Linux implementation, which is being developed at the University of Arizona, uses an object-based framework for implementing network protocols called x-kernel, and can be found at http://www.cs.arizona.edu/xkernel/hpcc-blue/linux.html. Most simply, the xkernel is a method of passing messages at the kernel level, which makes for an easier implementation.
David Jones (20.01.00) Page 452

85321, Systems Administration

Chapter 17: Security

Another freely-available IPSEC implementation is the Linux FreeS/WAN IPSEC. Their web page states, "These services allow you to build secure tunnels through untrusted networks. Everything passing through the untrusted net is encrypted by the IPSEC gateway machine and decrypted by the gateway at the other end. The result is Virtual Private Network or VPN. This is a network which is effectively private even though it includes machines at several diffeent sites connected by the insecure Internet." Its available for download from http://www.xs4all.nl/~freeswan/, and has just reached 1.0 at the time of this writing. As with other forms of cryptography, it is not distributed with the kernel by default due to export restrictions.

ssh (Secure Shell) and stelnet

ssh and stelnet are programs that allow you to login to remote systems and have a encrypted connection. ssh is a suite of programs used as a secure replacement for rlogin, rsh and rcp. It uses public-key cryptography to encrypt communications between two hosts, as well as to authenticate users. It can be used to securely login to a remote host or copy data between hosts, while preventing man-in-the-middle attacks (session hijacking) and DNS spoofing. It will perform data compression on your connections, and secure X11 communications between hosts. The ssh home page can be found at http://www.cs.hut.fi/ssh/ You can also use ssh from your Windows workstation to your Linux ssh server. There are several freely available Windows client implementations, including the one at http://guardian.htu.tuwien.ac.at/therapy/ssh/ as well as a commercial implementation from DataFellows, at http://www.datafellows.com. There is also a open source project to reimplement ssh called "psst...". For more information see: http://www.net.lut.ac.uk/psst/ SSLeay is a free implementation of Netscapes Secure Sockets Layer protocol, developed by Eric Young. It includes several applications, such as Secure telnet, a module for Apache, several databases, as well as several algorithms including DES, IDEA and Blowfish. Using this library, a secure telnet replacement has been created that does encryption over a telnet connection. Unlike SSH, stelnet uses SSL, the Secure Sockets Layer protocol developed by Netscape. You can find Secure telnet and Secure FTP by starting with the SSLeay FAQ, available at http://www.psy.uq.oz.au/~ftp/Crypto/. SRP is another secure telnet/ftp implementation. From their web page: "The SRP project is developing secure Internet software for free worldwide use. Starting with a fully-secure Telnet and FTP distribution, we hope to supplant weak networked authentication systems with strong replacements that do not sacrifice user-friendliness for security. Security should be the default, not an option!" For more information, go to http://srp.stanford.edu/srp.

David Jones (20.01.00)

Page 453

85321, Systems Administration

Chapter 17: Security

PAM - Pluggable Authentication Modules

Newer versions of the Red Hat Linux distribution ship with a unified authentication scheme called "PAM". PAM allows you to change your authentication methods and requirements on the fly, and encapsulate all local authentication methods without recompiling any of your binaries. Configuration of PAM is beyond the scope of this document, but be sure to take a look at the PAM web site for more information. http://www.kernel.org/pub/linux/libs/pam/index.html. Just a few of the things you can do with PAM: Use encryption other than DES for your passwords. (Making them harder to brute-force decode) Set resource limits on all your users so they cant perform denial-of-service attacks (number of processes, amount of memory, etc) Enable shadow passwords (see below) on the fly allow specific users to login only at specific times from specific places

Within a few hours of installing and configuring your system, you can prevent many attacks before they even occur. For example, use PAM to disable the system-wide usage of .rhosts files in users home directories by adding these lines to /etc/pam.d/rlogin:
# # Disable rsh/rlogin/rexec for users # login auth required pam_rhosts_auth.so no_rhosts

Cryptographic IP Encapsulation (CIPE)

The primary goal of this software is to provide a facility for secure (against eavesdropping, including traffic analysis, and faked message injection) subnetwork interconnection across an insecure packet network such as the Internet. CIPE encrypts the data at the network level. Packets traveling between hosts on the network are encrypted. The encryption engine is placed near the driver which sends and receives packets. This is unlike SSH, which encrypts the data by connection, at the socket level. A logical connection between programs running on different hosts is encrypted. CIPE can be used in tunnelling, in order to create a Virtual Private Network. Low-level encryption has the advantage that it can be made to work transparently between the two networks connected in the VPN, without any change to application software. Summarized from the CIPE documentation: The IPSEC standards define a set of protocols which can be used (among other things) to build encrypted VPNs. However, IPSEC is a rather heavyweight and complicated protocol set with a lot of options, implementations of the full protocol set are still rarely used and some issues (such as key management) are still not fully resolved. CIPE uses a simpler approach, in which many things which can be parameterized (such as the
David Jones (20.01.00) Page 454

85321, Systems Administration

Chapter 17: Security

choice of the actual encryption algorithm used) are an install-time fixed choice. This limits flexibility, but allows for a simple (and therefore efficient, easy to debug...) implementation. Further information can be found at http://www.inka.de/~bigred/devel/cipe.html As with other forms of cryptography, it is not distributed with the kernel by default due to export restrictions.

Kerberos
Kerberos is an authentication system developed by the Athena Project at MIT. When a user logs in, Kerberos authenticates that user (using a password), and provides the user with a way to prove her identity to other servers and hosts scattered around the network. This authentication is then used by programs such as rlogin to allow the user to login to other hosts without a password (in place of the .rhosts file). This authentication method can also used by the mail system in order to guarantee that mail is delivered to the correct person, as well as to guarantee that the sender is who he claims to be. Kerberos and the other programs that come with it, prevent users from "spoofing" the system into believing they are someone else. Unfortunately, installing Kerberos is very intrusive, requiring the modification or replacement of numerous standard programs. You can find more information about kerberos by looking at the kerberos FAQ, and the code can be found at http://nii.isi.edu/info/kerberos/. [From: Stein, Jennifer G., Clifford Neuman, and Jeffrey L. Schiller. "Kerberos: An Authentication Service for Open Network Systems." USENIX Conference Proceedings, Dallas, Texas, Winter 1998.] Kerberos should not be your first step in improving security of your host. It is quite involved, and not as widely used as, say, SSH.

Shadow Passwords.
Shadow passwords are a means of keeping your encrypted password information secret from normal users. Normally, this encrypted passwords are stored in /etc/passwd file for all to read. Anyone can then run password guesser programs on them and attempt to determine what they are. Shadow passwords, by contrast, are saved in /etc/shadow, which only privileged users can read. In order to use shadow passwords, you need to make sure all your utilities that need access to password information are recompiled to support them. PAM (above) also allows you to just plug in a shadow module; it doesnt require recompilation of executables. You can refer to the Shadow-Password HOWTO for further information if necessary. It is available at http://metalab.unc.edu/LDP/HOWTO/Shadow-Password-HOWTO.html It is rather dated now, and will not be required for distributions supporting PAM.

David Jones (20.01.00)

Page 455

85321, Systems Administration

Chapter 17: Security

Crack" and "John the Ripper"

If for some reason your passwd program is not enforcing hard-to-guess passwords, you might want to run a password-cracking program and make sure your users passwords are secure. Password cracking programs work on a simple idea: they try every word in the dictionary, and then variations on those words, encrypting each one and checking it against your encrypted password. If they get a match they know what your password is. There are a number of programs out there...the two most notable of which are "Crack" and "John the Ripper" (http://www.false.com/security/john/index.html) . They will take up a lot of your cpu time, but you should be able to tell if an attacker could get in using them by running them first yourself and notifying users with weak passwords. Note that an attacker would have to use some other hole first in order to read your /etc/passwd file, but such holes are more common than you might think. Because security is only as strong as the most insecure host, it is worth mentioning that if you have any Windows machines on your network, you should check out L0phtCrack, a Crack implementation for Windows. Its available from http://www.l0pht.com

CFS & TCFS: Cryptographic File Systems

CFS is a way of encrypting an entire directory trees and allowing users to store encrypted files on them. It uses a NFS server running on the local machine. RPMS are available at http://www.replay.com/redhat/, and more information on how it all works is at ftp://ftp.research.att.com/dist/mab/. TCFS improves on CFS by adding more integration with the file system, so that its transparent to users that the file system that is encrypted. more information at: http://edu-gw.dia.unisa.it/tcfs/. It also need not be used on entire filesystems. It works on directories trees as well.

X11, SVGA and display security

X11 Its important for you to secure your graphical display to prevent attackers from grabbing your passwords as you type them, reading documents or information you are reading on your screen, or even using a hole to gain root access. Running remote X applications over a network also can be fraught with peril, allowing sniffers to see all your interaction with the remote system. X has a number of access-control mechanisms. The simplest of them is hostbased: you use xhost to specify what hosts are allowed access to your display. This is not very secure at all, because if someone has access to your machine, they can xhost + their machine and get in easily. Also, if you have to allow access from an untrusted machine, anyone there can compromise your display. When using xdm (X Display Manager) to log in, you get a much better access method: MIT-MAGIC-COOKIE-1. A 128-bit "cookie" is generated and stored in your .Xauthority file. If you need to allow a remote machine access to your
David Jones (20.01.00) Page 456

85321, Systems Administration

Chapter 17: Security

display, you can use the xauth command and the information in your .Xauthority file to provide access to only that connection. See the Remote-XApps mini-howto, available at http://metalab.unc.edu/LDP/HOWTO/mini/Remote-X-Apps.html. You can also use ssh (see , above) to allow secure X connections. This has the advantage of also being transparent to the end user, and means that no unencrypted data flows across the network. Take a look at the Xsecurity man page for more information on X security. The safe bet is to use xdm to login to your console and then use ssh to go to remote sites on which you with to run X programs. SVGA SVGAlib programs are typically SUID-root in order to access all your Linux machines video hardware. This makes them very dangerous. If they crash, you typically need to reboot your machine to get a usable console back. Make sure any SVGA programs you are running are authentic, and can at least be somewhat trusted. Even better, dont run them at all. GGI (Generic Graphics Interface project) The Linux GGI project is trying to solve several of the problems with video interfaces on Linux. GGI will move a small piece of the video code into the Linux kernel, and then control access to the video system. This means GGI will be able to restore your console at any time to a known good state. They will also allow a secure attention key, so you can be sure that there is no Trojan horse login program running on your console. http://synergy.caltech.edu/~ggi/

Kernel Security
This is a description of the kernel configuration options that relate to security, and an explanation of what they do, and how to use them. As the kernel controls your computers networking, it is important that it be very secure, and not be compromised. To prevent some of the latest networking attacks, you should try to keep your kernel version current. You can find new kernels at <ftp://ftp.kernel.org> or from your distribution vendor. There is also a international group providing a single unified crypto patch to the mainstream linux kernel. This patch provides support for a number of cyrptographic subsystems and things that cannot be included in the mainstream kernel due to export restrictions. For more information, visit their web page at: http://www.kerneli.org

2.0 Kernel Compile Options

For 2.0.x kernels, the following options apply. You should see these options during the kernel configuration process. Many of the comments here are from ./linux/Documentation/Configure.help, which is the same document that is referenced while using the Help facility during the make config stage of compiling the kernel.
David Jones (20.01.00) Page 457

85321, Systems Administration

Chapter 17: Security

Network Firewalls (CONFIG_FIREWALL) This option should be on if you intend to run any firewalling or masquerading on your linux machine. If its just going to be a regular client machine, its safe to say no. IP: forwarding/gatewaying (CONFIG_IP_FORWARD) If you enable IP forwarding, your Linux box essentially becomes a router. If your machine is on a network, you could be forwarding data from one network to another, and perhaps subverting a firewall that was put there to prevent this from happening. Normal dial-up users will want to disable this, and other users should concentrate on the security implications of doing this. Firewall machines will want this enabled, and used in conjunction with firewall software. You can enable IP forwarding dynamically using the following command:
root# echo 1 > /proc/sys/net/ipv4/ip_forward

and disable it with the command:

root# echo 0 > /proc/sys/net/ipv4/ip_forward

Keep in mind the files, and their sizes, do not reflect their actual sizes, and despite being zero-length, may or may not be. IP: syn cookies (CONFIG_SYN_COOKIES) a "SYN Attack" is a denial of service (DoS) attack that consumes all the resources on your machine, forcing you to reboot. We cant think of a reason you wouldnt normally enable this. In the 2.1 kernel series this config option mearly allows syn cookies, but does not enable them. To enable them, you have to do:
root# echo 1 > /proc/sys/net/ipv4/tcp_syncookies

IP: Firewalling (CONFIG_IP_FIREWALL) This option is necessary if you are going to configure your machine as a firewall, do masquerading, or wish to protect your dial-up workstation from someone entering via your PPP dial-up interface. IP: firewall packet logging (CONFIG_IP_FIREWALL_VERBOSE) This option gives you information about packets your firewall received, like sender, recipient, port, etc. IP: Drop source routed frames (CONFIG_IP_NOSR) This option should be enabled. Source routed frames contain the entire path to their destination inside of the packet. This means that routers through which the packet goes do not need to inspect it, and just forward it on. This could lead to data entering your system that may be a potential exploit. IP: masquerading (CONFIG_IP_MASQUERADE) If one of the computers on your local network for which your Linux box acts as a firewall wants to send something to the outside, your box can "masquerade" as that host, i.e., it forwards the traffice to the intended destination, but makes it look like it came from the firewall box itself. See http://www.indyramp.com/masq for more information. IP: ICMP masquerading (CONFIG_IP_MASQUERADE_ICMP) This option adds ICMP masquerading to the previous option of only masquerading TCP or UDP traffic. IP: transparent proxy support (CONFIG_IP_TRANSPARENT_PROXY) This enables your Linux firewall to transparently redirect any network
Page 458

David Jones (20.01.00)

85321, Systems Administration

Chapter 17: Security

traffice originating from the local network and destined for a remote host to a local server, called a "transparent proxy server". This makes the local computers think they are talking to the remote end, while in fact they are connected to the local proxy. See the IP-Masquerading HOWTO and http://www.indyramp.com/masq for more information. IP: always defragment (CONFIG_IP_ALWAYS_DEFRAG) Generally this option is disabled, but if you are building a firewall or a masquerading host, you will want to enable it. When data is sent from one host to another, it does not always get sent as a single packet of data, but rather it is fragmented into several pieces. The problem with this is that the port numbers are only stored in the first fragment. This means that someone can insert information into the remaining packets that isnt supposed to be there. It could also prevent a teardrop attack against an internal host that is not yet itself patched against it. Packet Signatures (CONFIG_NCPFS_PACKET_SIGNING) This is an option that is available in the 2.1 kernel series that will sign NCP packets for stronger security. Normally you can leave it off, but it is there if you do need it. IP: Firewall packet netlink device (CONFIG_IP_FIREWALL_NETLINK) This is a really neat option that allows you to analyze the first 128 bytes of the packets in a user-space program, to determine if you would like to accept or deny the packet, based on its validity.

2.2 Kernel Compile Options

For 2.2.x kernels, many of the options are the same, but a few new ones have been developed. Many of the comments here are from ./linux/Documentation/Configure.help, which is the same document that is referenced while using the Help facility during the make config stage of compiling the kernel. Only the newly- added options are listed below. Consult the 2.0 description for a list of other necessary options. The most signficant change in the 2.2 kernel series is the IP firewalling code. The ipchains program is now used to install IP firewalling, instead of the ipfwadm program used in the 2.0 kernel. Socket Filtering (CONFIG_FILTER) For most people, its safe to say no to this option. This option allows you to connect a userspace filter to any socket and determine if packets should be allowed or denied. Unless you have a very specific need and are capable of programming such a filter, you should say no. Also note that as of this writing, all protocols were supported except TCP. Port Forwarding Port Forwarding is an addition to IP Masquerading which allows some forwarding of packets from outside to inside a firewall on given ports. This could be useful if, for example, you want to run a web server behind the firewall or masquerading host and that web server should be accessible from the outside world. An external client sends a request to port 80 of the firewall, the firewall forwards this request to the web server, the web server handles the request and the results are sent through the firewall to the original client. The client thinks that the firewall machine itself is running the web server. This can also be used for load balancing if you have a farm of identical web servers behind the firewall.
Page 459

David Jones (20.01.00)

85321, Systems Administration

Chapter 17: Security

Information about this feature is available from http://www.monmouth.demon.co.uk/ipsubs/portforwarding.html (to browse the WWW, you need to have access to a machine on the Internet that has a program like lynx or netscape). For general info, please see ftp://ftp.compsoc.net/users/steve/ipportfw/linux21/ Socket Filtering (CONFIG_FILTER) Using this option, user-space programs can attach a filter to any socket and thereby tell the kernel that it should allow or disallow certain types of data to get through the socket. Linux socket filtering works on all socket types except TCP for now. See the text file ./linux/Documentation/networking/filter.txt for more information. IP: Masquerading The 2.2 kernel masquerading has been improved. It provides additional support for masquerading special protocols, etc. Be sure to read the IP Chains HOWTO for more information.

Kernel Devices
There are a few block and character devices available on Linux that will also help you with security. The two devices /dev/random and /dev/urandom are provided by the kernel to provide random data at any time. Both /dev/random and /dev/urandom should be secure enough to use in generating PGP keys, ssh challenges, and other applications where secure random numbers are requisite. Attackers should be unable to predict the next number given any initial sequence of numbers from these sources. There has been a lot of effort put in to ensuring that the numbers you get from these sources are random in every sense of the word. The only difference is that /dev/random runs out of random bytes and it makes you wait for more to be accumulated. Note that on some systems, it can block for a long time waiting for new user-generated entry to be entered into the system. So you have to use care before using /dev/random. (Perhaps the best thing to do is to use it when youre generating sensitive keying information, and you tell the user to pound on the keyboard repeatedly until you print out "OK, enough".) /dev/random is high quality entropy, generated from measuring the interinterrupt times etc. It blocks until enough bits of random data are available. /dev/urandom is similar, but when the store of entropy is running low, itll return a cryptographically strong hash of what there is. This isnt as secure, but its enough for most applications. You might read from the devices using something like:
root# head -c 6 /dev/urandom | mmencode

This will print six random characters on the console, suitable for password generation. You can find mmencode in the metamail package. See /usr/src/linux/drivers/char/random.c for a description of the algorithm. Thanks to Theodore Y. Tso, Jon Lewis, and others from Linux-kernel for helping me (Dave) with this.

David Jones (20.01.00)

Page 460

85321, Systems Administration

Chapter 17: Security

Network Security
Network security is becoming more and more important as people spend more and more time connected. Compromising network security is often much easier than compromising physical or local, and is much more common. There are a number of good tools to assist with network security, and more and more of them are shipping with Linux distributions.

Packet Sniffers
One of the most common ways intruders gain access to more systems on your network is by employing a packet sniffer on a already compromised host. This "sniffer" just listens on the Ethernet port for things like passwd and login and su in the packet stream and then logs the traffic after that. This way, attackers gain passwords for systems they are not even attempting to break into. Cleartext passwords are very vulnerable to this attack. Example: Host A has been compromised. Attacker installs a sniffer. Sniffer picks up admin logging into Host B from Host C. It gets the admins personal password as they login to B. Then, the admin does a su to fix a problem. They now have the root password for Host B. Later the admin lets someone telnet from his account to Host Z on another site. Now the attacker has a password/login on Host Z. In this day and age, the attacker doesnt even need to compromise a system to do this: they could also bring a laptop or pc into a building and tap into your net. Using ssh or other encrypted password methods thwarts this attack. Things like APOP for POP accounts also prevents this attack. (Normal POP logins are very vulnerable to this, as is anything that sends clear-text passwords over the network.)

System services and tcp_wrappers

Before you put your Linux system on ANY network the first thing to look at is what services you need to offer. Services that you do not need to offer should be disabled so that you have one less thing to worry about and attackers have one less place to look for a hole. There are a number of ways to disable services under Linux. You can look at your /etc/inetd.conf file and see what services are being offered by your inetd. Disable any that you do not need by commenting them out (# at the beginning of the line), and then sending your inetd process a SIGHUP. You can also remove (or comment out) services in your /etc/services file. This will mean that local clients will also be unable to find the service (i.e., if you remove ftp, and try and ftp to a remote site from that machine it will fail with an "unknown service" message). Its usually not worth the trouble to remove services, since it provides no additional security. If a local person wanted to use ftp even though you had commented it out, they would make their own client that use the common FTP port and would still work fine. Some of the services you might want to leave enabled are: ftp
Page 461 David Jones (20.01.00)

85321, Systems Administration

Chapter 17: Security

telnet (or ssh) mail, such as pop-3 or imap identd

If you know you are not going to use some particular package, you can also delete it entirely. rpm -e packagename under the Red Hat distribution will erase an entire package. Under debian dpkg --remove does the same thing. Additionally, you really want to disable the rsh/rlogin/rcp utilities, including login (used by rlogin), shell (used by rcp), and exec (used by rsh) from being started in /etc/inetd.conf. These protocols are extremely insecure and have been the cause of exploits in the past. You should check your /etc/rc.d/rcN.d, (where N is your systems run level) and see if any of the servers started in that directory are not needed. The files in /etc/rc.d/rcN.d are actually symbolic links to the directory /etc/rc.d/init.d. Renaming the files in the init.d directory has the effect of disabling all the symbolic links in /etc/rc.d/rcN.d. If you only wish to disable a service for a particular run level, rename the appropriate file by replacing the upper-case S with a lower-case s, like this:
root# root# cd /etc/rc6.d mv S45dhcpd s45dhcpd

If you have BSD style rc files, you will want to check /etc/rc* for programs you dont need. Most Linux distributions ship with tcp_wrappers "wrapping" all your TCP services. A tcp_wrapper (tcpd) is invoked from inetd instead of the real server. tcpd then checks the host that is requesting the service, and either executes the real server, or denies access from that host. tcpd allows you to restrict access to your TCP services. You should make a /etc/hosts.allow and add in only those hosts that need to have access to your machines services. If you are a home dialup user, we suggest you deny ALL. tcpd also logs failed attempts to access services, so this can give alert you if you are under attack. If you add new services, you should be sure to configure them to use tcp_wrappers if they are TCP based. For example, a normal dial-up user can prevent outsiders from connecting to his machine, yet still have the ability to retrieve mail, and make network connections to the Internet. To do this, you might add the following to your /etc/hosts.allow:
ALL: 127.

And of course /etc/hosts.deny would contain:

ALL: ALL

which will prevent external connections to your machine, yet still allow you from the inside to connect to servers on the Internet. Keep in mind that tcp_wrappers only protect services executed from inetd, and a select few others. There very well may be other services running on your machine. You can use netstat -ta to find a list of all the services your machine is offering.

Verify Your DNS Information

Keeping up-to-date DNS information about all hosts on your network can help to increase security. If an unauthorized host becomes connected to your
David Jones (20.01.00) Page 462

85321, Systems Administration

Chapter 17: Security

network, you can recognize it by its lack of a DNS entry. Many services can be configured to not accept connections from hosts that do not have valid DNS entries.

identd
identd is a small program that typically runs out of your inetd server. It keeps track of what user is running what TCP service, and then reports this to whoever requests it. Many people misunderstand the usefulness of identd, and so disable it or block all off site requests for it. identd is not there to help out remote sites. There is no way of knowing if the data you get from the remote identd is correct or not. There is no authentication in identd requests. Why would you want to run it then? Because it helps you out, and is another data-point in tracking. If your identd is un compromised, then you know its telling remote sites the user-name or uid of people using TCP services. If the admin at a remote site comes back to you and tells you user so-and-so was trying to hack into their site, you can easily take action against that user. If you are not running identd, you will have to look at lots and lots of logs, figure out who was on at the time, and in general take a lot more time to track down the user. The identd that ships with most distributions is more configurable than many people think. You can disable it for specific users (they can make a .noident file), you can log all identd requests (We recommend it), you can even have identd return a uid instead of a user name or even NO-USER.

SATAN, ISS, and Other Network Scanners

There are a number of different software packages out there that do port and service based scanning of machines or networks. SATAN, ISS, SAINT, and Nessus are some of the more well-known ones. This software connects to the target machine (or all the target machines on a network) on all the ports they can, and try to determine what service is running there. Based on this information, you can tell if the machine is vulnerable to a specific exploit on that server. SATAN (Security Administrators Tool for Analyzing Networks) is a port scanner with a web interface. It can be configured to do light, medium, or strong checks on a machine or a network of machines. Its a good idea to get SATAN and scan your machine or network, and fix the problems it finds. Make sure you get the copy of SATAN from metalab <http://metalab.unc.edu/pub/packages/security/Satan-for-Linux/> or a reputable FTP or web site. There was a Trojan copy of SATAN that was distributed out on the net. http://www.trouble.org/~zen/satan/satan.html. Note that SATAN has not been updated in quite a while, and some of the other tools below might do a better job. ISS (Internet Security Scanner) is another port-based scanner. It is faster than Satan, and thus might be better for large networks. However, SATAN tends to provide more information.

David Jones (20.01.00)

Page 463

85321, Systems Administration

Chapter 17: Security

Abacus is a suite of tools to provide host based security and intrusion detection. look at its home page on the web for more information. http://www.psionic.com/abacus/ SAINT is a updated version of SATAN. It is web based and has many more up to date tests than SATAN. You can find out more about it at: http://www.wwdsi.com/~saint Nessus is a free security scanner. It has a GTK graphical interface for ease of use. It is also designed with a very nice plugin setup for new port scanning tests. For more information, take a look at: http://www.nessus.org

Detecting Port Scans

There are some tools designed to alert you to probes by SATAN and ISS and other scanning software. However, liberal use of tcp_wrappers, and make sure to look over your log files regularly, you should be able to notice such probes. Even on the lowest setting, SATAN still leaves traces in the logs on a stock Red Hat system. There are also "stealth" port scanners. A packet with the TCP ACK bit set (as is done with established connections) will likely get through a packet-filtering firewall. The returned RST packet from a port that _had no established session_ can be taken as proof of life on that port. I dont think TCP wrappers will detect this.

sendmail , qmail and MTAs

One of the most important services you can provide is a mail server. Unfortunately, it is also one of the most vulnerable to attack, simply due to the number of tasks it must perform and the privileges it typically needs. If you are using sendmail it is very important to keep up on current versions. sendmail has a long long history of security exploits. Always make sure you are running the most recent version from http://www.sendmail.org. Keep in mind that sendmail does not have to be running in order for you to send mail. If you are a home user, you can disable sendmail entirely, and simply use your mail client to send mail. You might also choose to remove the "-bd" flag from the sendmail startup file, thereby disabling incoming requests for mail. In other words, you can execute sendmail from your startup script using the following instead:
# /usr/lib/sendmail -q15m

This will cause sendmail to flush the mail queue every fifteen minutes for any messages that could not be successfully delivered on the first attempt. Many administrators choose not to use sendmail, and instead choose one of the other mail transport agents. You might consider switching over to qmail. qmail was designed with security in mind from the ground up. Its fast, stable, and secure. Qmail can be found at http://www.qmail.org In direct competition to qmail is "postfix", written by Wietse Venema, the author of tcp_wrappers and other security tools. Formerly called vmailer, and sponsored by IBM, this is also a mail transport agent written from the ground up with security in mind. You can find more information about vmailer at http://www.postfix.org
David Jones (20.01.00) Page 464

85321, Systems Administration

Chapter 17: Security

Denial of Service Attacks

A "Denial of Service" (DoS) attack is one where the attacker tries to make some resource too busy to answer legitimate requests, or to deny legitimate users access to your machine. Denial of service attacks have increased greatly in recent years. Some of the more popular and recent ones are listed below. Note that new ones show up all the time, so this is just a few examples. Read the Linux security lists and the bugtraq list and archives for more current information. SYN Flooding - SYN flooding is a network denial of service attack. It takes advantage of a "loophole" in the way TCP connections are created. The newer Linux kernels (2.0.30 and up) have several configurable options to prevent SYN flood attacks from denying people access to your machine or services. See Kernel Security for proper kernel protection options. Pentium "F00F" Bug - It was recently discovered that a series of assembly codes sent to a genuine Intel Pentium processor would reboot the machine. This affects every machine with a Pentium processor (not clones, not Pentium Pro or PII), no matter what operating system its running. Linux kernels 2.0.32 and up contain a work around for this bug, preventing it from locking your machine. Kernel 2.0.33 has an improved version of the kernel fix, and is suggested over 2.0.32. If you are running on a Pentium, you should upgrade now! Ping Flooding - Ping flooding is a simple brute-force denial of service attack. The attacker sends a "flood" of ICMP packets to your machine. If they are doing this from a host with better bandwidth than yours, your machine will be unable to send anything on the network. A variation on this attack, called "smurfing", sends ICMP packets to a host with your machines return IP, allowing them to flood you less detectably. You can find more information about the "smurf" attack at http://www.quadrunner.com/~chuegen/smurf.txt If you are ever under a ping flood attack, use a tool like tcpdump to determine where the packets are coming from (or appear to be coming from), then contact your provider with this information. Ping floods can most easily be stopped at the router level or by using a firewall. Ping o Death - The Ping o Death attack sends ICMP ECHO REQUEST packets that are too large to fit in the kernel data structures intended to store them. Because sending a single, large (65,510 bytes) "ping" packet to many systems will cause them to hang or even crash, this problem was quickly dubbed the "Ping o Death." This one has long been fixed, and is no longer anything to worry about. Teardrop / New Tear - One of the most recent exploits involves a bug present in the IP fragmentation code on Linux and Windows platforms. It is fixed in kernel version 2.0.33, and does not require selecting any kernel compile-time options to utilize the fix. Linux is apparently not vulnerable to the "newtear" exploit.

You can find code for most exploits, and a more in-depth description of how they work, at http://www.rootshell.com using their search engine.
David Jones (20.01.00) Page 465

85321, Systems Administration

Chapter 17: Security

NFS (Network File System) Security.

NFS is a very widely-used file sharing protocol. It allows servers running nfsd and mountd to "export" entire filesystems to other machines using NFS filesystem support built in to their kernels (or some other client support if they are not Linux machines). mountd keeps track of mounted filesystems in /etc/mtab, and can display them with showmount. Many sites use NFS to serve home directories to users, so that no matter what machine in the cluster they login to, they will have all their home files. There is some small amount of security allowed in exporting filesystems. You can make your nfsd map the remote root user (uid=0) to the nobody user, denying them total access to the files exported. However, since individual users have access to their own (or at least the same uid) files, the remote root user can login or su to their account and have total access to their files. This is only a small hindrance to an attacker that has access to mount your remote filesystems. If you must use NFS, make sure you export to only those machines that you really need to. Never export your entire root directory; export only directories you need to export. See the NFS HOWTO for more information on NFS, available at http://metalab.unc.edu/mdw/HOWTO/NFS-HOWTO.html

NIS (Network Information Service) (formerly YP).

Network Information service (formerly YP) is a means of distributing information to a group of machines. The NIS master holds the information tables and converts them into NIS map files. These maps are then served over the network, allowing NIS client machines to get login, password, home directory and shell information (all the information in a standard /etc/passwd file). This allows users to change their password once and have it take effect on all the machines in the NIS domain. NIS is not at all secure. It was never meant to be. It was meant to be handy and useful. Anyone that can guess the name of your NIS domain (anywhere on the net) can get a copy of your passwd file, and use "crack" and "John the Ripper" against your users passwords. Also, it is possible to spoof NIS and do all sorts of nasty tricks. If you must use NIS, make sure you are aware of the dangers. There is a much more secure replacement for NIS, called NIS+. Check out the NIS HOWTO for more information: http://metalab.unc.edu/mdw/HOWTO/NIS-HOWTO.html

Firewalls
Firewalls are a means of controlling what information is allowed into and out of your local network. Typically the firewall host is connected to the Internet and your local LAN, and the only access from your LAN to the Internet is through the firewall. This way the firewall can control what passes back and forth from the Internet and your lan. There are a number of types of firewalls and methods of setting them up. Linux machines make pretty good firewalls. Firewall code can be built right into 2.0 and higher kernels. The ipfwadm for 2.0 kernels, or ipchains for 2.2
David Jones (20.01.00) Page 466

85321, Systems Administration

Chapter 17: Security

kernels, user-space tools allows you to change, on the fly, the types of network traffic you allow. You can also log particular types of network traffic. Firewalls are a very useful and important technique in securing your network. However, never think that because you have a firewall, you dont need to secure the machines behind it. This is a fatal mistake. Check out the very good Firewall-HOWTO at your latest metalab archive for more information on firewalls and Linux. http://metalab.unc.edu/mdw/HOWTO/FirewallHOWTO.html More information can also be found in the IP-Masquerade mini-howto: http://metalab.unc.edu/mdw/HOWTO/mini/IP-Masquerade.html More information on ipfwadm (The tool that lets you change settings on your firewall, can be found at its home page: http://www.xos.nl/linux/ipfwadm/ If you have no experience with firewalls, and plan to set up one for more than just a simple security policy, the Firewalls book by OReilly and Associates or other online firewall document is mandatory reading. Check out http://www.ora.com for more information. The National Institute of Standards and Technology have put together an excellent document on firewalls. Although dated 1995, it is still quite good. You can find it at http://csrc.nist.gov/nistpubs/800-10/main.html. Also of interest includes: The Freefire Project -- a list of freely-available firewall tools, available at http://sites.inka.de/sites/lina/freefire-l/index_en.html SunWorld Firewall Design -- written by the authors of the OReilly book, this provides a rough introduction to the different firewall types. Its available at http://www.sunworld.com/swol-01-1996/swol-01-firewall.html

IP Chains - Linux Kernel 2.2.x Firewalling

Linux IP Firewalling Chains is an update to the 2.0 Linux firewalling code for the 2.2 kernel. It has a great deal more features than previous implementations, including: More flexible packet manipulations More complex accounting Simple policy changes possible atomically Fragments can be explicitly blocked, denied, etc. Logs suspicious packets. Can handle protocols other than ICMP/TCP/UDP.

If you are currently using ipfwadm on your 2.0 kernel, there are scripts available to convert the ipfwadm command format to the format ipchains uses. Be sure to read the IP Chains HOWTO for further information. It is avilable at http://www.rustcorp.com/linux/ipchains/HOWTO.html

VPNs - Virtual Private Networks

VPNs are a way to establish a "virtual" network on top of some already existing network. This virtual network often is encrypted and passes traffic only to and from some known entities that have joined the network. VPNs are
David Jones (20.01.00) Page 467

85321, Systems Administration

Chapter 17: Security

often used to connect someone working at home over the public internet to a internal company network by using a encrypted virtual network. If you are running a linux masquerading firewall and need to pass MS PPTP (Microsofts VPN point to point product) packets, there is a linux kernel patch out to do just that. See: ip-masq-vpn. There are several linux VPN solutions available: vpnd. See the http://www.crosswinds.net/nuremberg/~anstein/unix/vpnd.html. Free S/Wan, available at http://www.xs4all.nl/~freeswan/ ssh can be used to construct a VPN. See the VPN mini-howto for more information. vps (virtual private server) at http://www.strongcrypto.com.

See also the section on IPSEC for pointers and more information.

Security Preparation (before you go on-line)

Ok, so you have checked over your system, and determined its as secure as feasible, and youre ready to put it online. There are a few things you should now do in order to prepare for an intrusion, so you can quickly disable the intruder, and get back up and running.

Make a Full Backup of Your Machine

Discussion of backup methods and storage is beyond the scope of this document, but here are a few words relating to backups and security: If you have less than 650mb of data to store on a partition, a CD-R copy of your data is a good way to go (as its hard to tamper with later, and if stored properly can last a long time). Tapes and other re-writable media should be write-protected as soon as your backup is complete, and then verified to prevent tampering. Make sure you store your backups in a secure off-line area. A good backup will ensure that you have a known good point to restore your system from.

Choosing a Good Backup Schedule

A six-tape cycle is easy to maintain. This includes four tapes for during the week, one tape for even Fridays, and one tape for odd Fridays. Perform an incremental backup every day, and a full backup on the appropriate Friday tape. If you make some particularly important changes or add some important data to your system, a full backup might well be in order.

Backup Your RPM or Debian File Database

In the event of an intrusion, you can use your RPM database like you would use tripwire, but only if you can be sure it too hasnt been modified. You should copy the RPM database to a floppy, and keep this copy off-line at all times. The Debian distribution likely has something similar.

David Jones (20.01.00)

Page 468

85321, Systems Administration

Chapter 17: Security

The files /var/lib/rpm/fileindex.rpm and /var/lib/rpm/packages.rpm most likely wont fit on a single floppy. But if Compressed, each should fit on a seperate floppy. Now, when your system is compromised, you can use the command:
root# rpm -Va

to verify each file on the system. See the rpm man page, as there are a few other options that can be included to make it less verbose. Keep in mind you must also be sure your RPM binary has not been compromised. This means that every time a new RPM is added to the system, the RPM database will need to be rearchived. You will have to decide the advantages versus drawbacks.

Keep Track of Your System Accounting Data

It is very important that the information that comes from syslog has not been compromised. Making the files in /var/log readable and writable by only a limited number of users is a good start. Be sure to keep an eye on what gets written there, especially under the auth facility. Multiple login failures, for example, can indicate an attempted breakin. Where to look for your log file will depend on your distribution. In a Linux system that conforms to the "Linux Filesystem Standard", such as Red Hat, you will want to look in /var/log and check messages, mail.log, and others. You can find out where your distribution is logging to by looking at your /etc/syslog.conf file. This is the file that tells syslogd (the system logging daemon) where to log various messages. You might also want to configure your log-rotating script or daemon to keep logs around longer so you have time to examine them. Take a look at the logrotate package on recent Red Hat distributions. Other distributions likely have a similar process. If your log files have been tampered with, see if you can determine when the tampering started, and what sort of things appeared to be tampered with. Are there large periods of time that cannot be accounted for? Checking backup tapes (if you have any) for untampered log files is a good idea. Log files are typically modified by the intruder in order to cover his tracks, but they should still be checked for strange happenings. You may notice the intruder attempting to gain entrance, or exploit a program in order to obtain the root account. You might see log entries before the intruder has time to modify them. You should also be sure to seperate the auth facility from other log data, including attempts to switch users using su, login attempts, and other user accounting information. If possible, configure syslog to send a copy of the most important data to a secure system. This will prevent an intruder from covering his tracks by deleting his login/su/ftp/etc attempts. See the syslog.conf man page, and refer to the @ option.

David Jones (20.01.00)

Page 469

85321, Systems Administration

Chapter 17: Security

There are several more advanced syslogd programs out there. Take a look at http://www.core-sdi.com/ssyslog/ for Secure Syslog. Secure Syslog allows you to encrypt your syslog entries and make sure no one has tampered with them. Another syslogd with more features is syslog-ng. It allows you a lot more flexability in your logging and also can has your remote syslog streams to prevent tampering. Finally, log files are much less useful when no one is reading them. Take some time out every once in a while to look over your log files, and get a feeling for what they look like on a normal day. Knowing this can help make unusual things stand out.

Apply All New System Updates.

Most Linux users install from a CD-ROM. Due to the fast-paced nature of security fixes, new (fixed) programs are always being released. Before you connect your machine to the network, its a good idea to check with your distributions ftp site and get all the updated packages since you received your distribution CD-ROM. Many times these packages contain important security fixes, so its a good idea to get them installed.

What To Do During and After a Breakin

So you have followed some of the advice here (or elsewhere) and have detected a break-in? The first thing to do is to remain calm. Hasty actions can cause more harm than the attacker would have.

Security Compromise Underway.

Spotting a security compromise under way can be a tense undertaking. How you react can have large consequences. If the compromise you are seeing is a physical one, odds are you have spotted someone who has broken into your home, office or lab. You should notify your local authorities. In a lab, you might have spotted someone trying to open a case or reboot a machine. Depending on your authority and procedures, you might ask them to stop, or contact your local security people. If you have detected a local user trying to compromise your security, the first thing to do is confirm they are in fact who you think they are. Check the site they are logging in from. Is it the site they normally log in from? No? Then use a non-electronic means of getting in touch. For instance, call them on the phone or walk over to their office/house and talk to them. If they agree that they are on, you can ask them to explain what they were doing or tell them to cease doing it. If they are not on, and have no idea what you are talking about, odds are this incident requires further investigation. Look into such incidents , and have lots of information before making any accusations. If you have detected a network compromise, the first thing to do (if you are able) is to disconnect your network. If they are connected via modem, unplug the modem cable; if they are connected via ethernet, unplug the Ethernet cable. This will prevent them from doing any further damage, and they will probably see it as a network problem rather than detection.
David Jones (20.01.00) Page 470

85321, Systems Administration

Chapter 17: Security

If you are unable to disconnect the network (if you have a busy site, or you do not have physical control of your machines), the next best step is to use something like tcp_wrappers or ipfwadm to deny access from the intruders site. If you cant deny all people from the same site as the intruder, locking the users account will have to do. Note that locking an account is not an easy thing. You have to keep in mind .rhosts files, FTP access, and a host of possible backdoors). After you have done one of the above (disconnected the network, denied access from their site, and/or disabled their account), you need to kill all their user processes and log them off. You should monitor your site well for the next few minutes, as the attacker will try to get back in. Perhaps using a different account, and/or from a different network address.

Security Compromise has already happened

So you have either detected a compromise that has already happened or you have detected it and locked (hopefully) the offending attacker out of your system. Now what? Closing the Hole If you are able to determine what means the attacker used to get into your system, you should try to close that hole. For instance, perhaps you see several FTP entries just before the user logged in. Disable the FTP service and check and see if there is an updated version, or if any of the lists know of a fix. Check all your log files, and make a visit to your security lists and pages and see if there are any new common exploits you can fix. You can find Caldera security fixes at http://www.caldera.com/tech-ref/security/. Red Hat has not yet seperated their security fixes from bug fixes, but their distribution errata is available at http://www.redhat.com/errata Debian now has a security mailing list and web page. See: http://www.debian.com/security/ for more information. It is very likely that if one vendor has released a security update, that most other Linux vendors will as well. There is now a linux security auditing project. They are methodically going through all the user space utilities and looking for possible security exploits and overflows. From their announcement: "We are attempting a systematic audit of Linux sources with a view to being as secure as OpenBSD. We have already uncovered (and fixed) some problems, but more help is welcome. The list is unmoderated and also a useful resource for general security discussions. The list address is: [email protected] To subscribe, send a mail to: [email protected]" If you dont lock the attacker out, they will likely be back. Not just back on your machine, but back somewhere on your network. If they were running a packet sniffer, odds are good they have access to other local machines.
David Jones (20.01.00) Page 471

85321, Systems Administration

Chapter 17: Security

Assessing the Damage The first thing is to assess the damage. What has been compromised? If you are running an Integrity Checker like Tripwire, you can use it to perform an integrity check, and should help to tell you. If not, you will have to look around at all your important data. Since Linux systems are getting easier and easier to install, you might consider saving your config files and then wiping your disk(s) and reinstalling, then restoring your user files from backups and your config files. This will ensure that you have a new, clean system. If you have to backup files from the compromised system, be especially cautious of any binaries that you restore, as they may be Trojan horses placed there by the intruder. Re-installation should be considered mandatory upon an intruder obtaining root access. Additionally, youd like to keep any evidence there is, so having a spare disk in the safe may make sense. Then you have to worry about how long ago the compromise happened, and whether the backups hold any damaged work. More on backups later. Backups, Backups, Backups! Having regular backups is a godsend for security matters. If your system is compromised, you can restore the data you need from backups. Of course, some data is valuable to the attacker too, and they will not only destroy it, they will steal it and have their own copies; but at least you will still have the data. You should check several backups back into the past before restoring a file that has been tampered with. The intruder could have compromised your files long ago, and you could have made many successful backups of the compromised file!!! Of course, there are also a raft of security concerns with backups. Make sure you are storing them in a secure place. Know who has access to them. (If an attacker can get your backups, they can have access to all your data without you ever knowing it.) Tracking Down the Intruder. Ok, you have locked the intruder out, and recovered your system, but youre not quite done yet. While it is unlikely that most intruders will ever be caught, you should report the attack. You should report the attack to the admin contact at the site where the attacker attacked your system. You can look up this contact with whois or the Internic database. You might send them an email with all applicable log entries and dates and times. If you spotted anything else distinctive about your intruder, you might mention that too. After sending the email, you should (if you are so inclined) follow up with a phone call. If that admin in turn spots your attacker, they might be able to talk to the admin of the site where they are coming from and so on. Good crackers often use many intermediate systems, some (or many) of which may not even know they have been compromised. Trying to track a cracker back to their home system can be difficult. Being polite to the admins you talk to can go a long way to getting help from them. You should also notify any
David Jones (20.01.00) Page 472

85321, Systems Administration

Chapter 17: Security

security organizations you are a part of (CERT <http://www.cert.org/> or similar), as well as your Linux system vendor.

Security Sources
There are a LOT of good sites out there for Unix security in general and Linux security specifically. Its very important to subscribe to one (or more) of the security mailing lists and keep current on security fixes. Most of these lists are very low volume, and very informative.

FTP Sites
CERT is the Computer Emergency Response Team. They often send out alerts of current attacks and fixes. See ftp://ftp.cert.org for more information. Replay (http://www.replay.com) has archives of many security programs. Since they are outside the US, they dont need to obey US crypto restrictions. Matt Blaze is the author of CFS and a great security advocate. Matts archive is available at ftp://ftp.research.att.com/pub/mab <ftp://ftp.research.att.com/pub/mab> tue.nl is a great security FTP site in the Netherlands. ftp.win.tue.nl

Web Sites
The Hacker FAQ is a FAQ about hackers: http://www.tuxedo.org/~esr/faqs/hacker-howto.html The COAST archive has a large number of Unix security programs and information: http://www.cerias.purdue.edu/coast/archive/index.html SuSe Security Page: http://www.suse.de/security/ Rootshell.com is a great site for seeing what exploits are currently being used by crackers: http://www.rootshell.com/ BUGTRAQ puts out advisories on security issues: http://www.securityfocus.com/forums/bugtraq/intro.html CERT, the Computer Emergency Response Team, puts out advisories on common attacks on unix platforms: http://www.cert.org/ Dan Farmer is the author of SATAN and many other security tools. His home site has some interesting security survey information, as well as security tools: http://www.trouble.org The Linux security WWW is a good site for Linux security information: http://www.ecst.csuchico.edu/~jtmurphy/ Infilsec has a vulnerability engine that can tell you what vunerabilities affect a specific platform: http://www.infilsec.com/vulnerabilities/ CIAC sends out periodic security bulletins on common exploits: http://ciac.llnl.gov/cgi-bin/index/bulletins A good starting point for Linux Pluggable Authentication modules can be found at http://www.kernel.org/pub/linux/libs/pam/. The debian project
Page 473

David Jones (20.01.00)

85321, Systems Administration

Chapter 17: Security

has a web page for their security fixes and information. It is at http://www.debian.com/security/. WWW Security FAQ, written by Lincoln Stein, is a great web security reference. Find it at http://www.w3.org/Security/Faq/www-securityfaq.html

Mailing Lists
Bugtraq: To subscribe to bugtraq, send mail to [email protected] containing the message body subscribe bugtraq. (see links above for archives). CIAC: Send e-mail to [email protected]. In the BODY (not subject) of the message put (either or both): subscribe ciac-bulletin Red Hat has a number of mailing lists, the most important of which is the redhat-announce list. You can read about security (and other) fixes as soon as they come out. Send email to [email protected] and put subscribe redhat-announce. The Debian project has a security mailing list that covers their security fixes. see http://www.debian.com/security/ for more information.

Books - Printed Reading Material

There are a number of good security books out there. This section lists a few of them. In addition to the security specific books, security is covered in a number of other books on system administration. Building Internet Firewalls By D. Brent Chapman & Elizabeth D. Zwicky 1st Edition September 1995 ISBN: 1-56592-124-0 Practical UNIX & Internet Security, 2nd Edition By Simson Garfinkel & Gene Spafford, 2nd Edition April 1996 ISBN: 1-56592-148-8 Computer Security Basics By Deborah Russell & G.T. Gangemi, Sr. 1st Edition July 1991 ISBN: 0-937175-71-4 Linux Network Administrators Guide By Olaf Kirch 1st Edition January 1995 ISBN: 1-56592-087-2 PGP: Pretty Good Privacy By Simson Garfinkel 1st Edition December 1994 ISBN: 1-56592-098-8 Computer Crime A Crimefighters Handbook By David Icove, Karl Seger & William VonStorch (Consulting Editor Eugene H. Spafford) 1st Edition August 1995 ISBN: 1-56592-086-4

Glossary
authentication: The property of knowing that the data received is the same as the data that was sent, and that the claimed sender is in fact the actual sender. bastion Host: A computer system that must be highly secured because it is vulnerable to attack, usually because it is exposed to the Internet and is a main point of contact for users of internal networks. It gets its name from the highly fortified projects on the outer walls of medieval castles. Bastions overlook critical areas of defense, usually having strong walls,
Page 474

David Jones (20.01.00)

85321, Systems Administration

Chapter 17: Security

room for extra troops, and the occasional useful tub of boiling hot oil for discouraging attackers. buffer overflow: Common coding style is to never allocate large enough buffers, and to not check for overflows. When such buffers overflow, the executing program (daemon or set-uid program) can be tricked in doing some other things. Generally this works by overwriting a functions return address on the stack to point to another location. denial of service: A denial of service attack is when an attacker consumes the resources on your computer for things it was not intended to be doing, thus preventing normal use of your network resources for legimite purposes. dual-homed Host: A general-purpose computer system that has at least two network interfaces. firewall: A component or set of components that restricts access between a protected network and the Internet, or between other sets of networks. host: A computer system attached to a network. IP spoofing: IP Spoofing is a complex technical attack that is made up of several components. It is a security exploit that works by tricking computers in a trust-relationship that you are someone that you really arent. There is an extensive paper written by daemon9, route, and infinity in the Volume Seven, Issue fourty-Eight issue of Phrack Magazine. non-repudiation: The property of a receiver being able to prove that the sender of some data did in fact send the data even though the sender might later deny ever having sent it. packet: The fundamental unit of communication on the Internet. packet filtering: The action a device takes to selectively control the flow of data to and from a network. Packet filters allow or block packets, usually while routing them from one network to another (most often from the Internet to an internal network, and vice-versa). To accomplish packet filtering, you set up rules that specify what types of packets (those to or from a particular IP address or port) are to be allowed and what types are to be blocked. perimeter network: A network added between a protected network and an external network, in order to provide an additional layer of security. A perimeter network is sometimes called a DMZ. proxy server: A program that deals with external servers on behalf of internal clients. Proxy clients talk to proxy servers, which relay approved client requests to real servers, and relay answers back to clients. superuser: An informal name for root.

Frequently Asked Questions

1. Is it more secure to compile driver support directly into the kernel, instead of making it a module? Answer: Some people think it is better to disable the ability to load device drivers using modules, because an intruder could load a Trojan module or a module that could affect system security.
David Jones (20.01.00) Page 475

85321, Systems Administration

Chapter 17: Security

However, in order to load modules, you must be root. The module object files are also only writable by root. This means the intruder would need root access to insert a module. If the intruder gains root access, there are more serious things to worry about than whether he will load a module. Modules are for dynamically loading support for a particular device that may be infrequently used. On server machines, or firewalls for instance, this is very unlikely to happen. For this reason, it would make more sense to compile support directly into the kernel for machines acting as a server. Modules are also slower than support compiled directly in the kernel. 2. Why does logging in as root from a remote machine always fail? Answer: See Root Security. This is done intentionally to prevent remote users from attempting to connect via telnet to your machine as root, which is a serious security vulnerability. Dont forget: potential intruders have time on their side, and can run automated programs to find your password. 3. How do I enable shadow passwords on my Red Hat 4.2 or 5.x Linux box? Answer: Shadow passwords is a mechanism for storing your password in a file other than the normal /etc/passwd file. This has several advantages. The first one is that the shadow file, /etc/shadow, is only readable by root, unlike /etc/passwd, which must remain readable by everyone. The other advantage is that as the administrator, you can enable or disable accounts without everyone knowing the status of other users accounts. The /etc/passwd file is then used to store user and group names, used by programs like /bin/ls to map the user ID to the proper username in a directory listing. The /etc/shadow file then only contains the username and his/her password, and perhaps accounting information, like when the account expires, etc. To enable shadow passwords, run pwconv as root, and /etc/shadow should now exist, and be used by applications. Since you are using RH 4.2 or above, the PAM modules will automatically adapt to the change from using normal /etc/passwd to shadow passwords without any other change. Since youre interested in securing your passwords, perhaps you would also be interested in generating good passwords to begin with. For this you can use the pam_cracklib module, which is part of PAM. It runs your password against the Crack libraries to help you decide if it is too easily guessable by password cracking programs. 4. How can I enable the Apache SSL extensions? Answer: 1.Get SSLeay 0.8.0 or later from ftp://ftp.psy.uq.oz.au/pub/Crypto/SSL 2.Build and test and install it! 3.Get Apache 1.2.5 source 1.3.10 is about to be released 4.Get Apache SSLeay extensions from here ftp://ftp.ox.ac.uk/pub/crypto/SSL/apache_1.2.5+ssl_1.13.tar.gz 5.Unpack it in the apache-1.2.5 source directory and patch Apache as per the README. 6.Configure and build it.
David Jones (20.01.00) Page 476

85321, Systems Administration

Chapter 17: Security

You might also try Replay Associates which has many pre-built packages, and is located outside of the United States. 5. How can I manipulate user accounts, and still retain security? Answer: The Red Hat distribution, especially RH5.0, contains a great number of tools to change the properties of user accounts. The pwconv and unpwconv programs can be used to convert between shadow and non-shadowed passwords. The pwck and grpck programs can be used to verify proper organization of the passwd and group files. The useradd, usermod, and userdel programs can be used to add, delete and modify user accounts. The groupadd, groupmod, and groupdel programs will do the same for groups. Group passwords can be created using gpasswd. All these programs are "shadow-aware" -- that is, if you enable shadow they will use /etc/shadow for password information, otherwise it wont. See the respective man pages for further information. 6. How can I password protect specific HTML documents using Apache? I bet you didnt know about http://www.apacheweek.org, did you? You can find information on user Authentication at http://www.apacheweek.com/features/userauth as well as other web server security tips from http://www.apache.org/docs/misc/security_tips.html

Conclusion
By subscribing to the security alert mailing lists, and keeping current, you can do a lot towards securing your machine. If you pay attention to your log files and run something like tripwire regularly, you can do even more. A reasonable level of computer security is not difficult to maintain on a home machine. More effort is required on business machines, but Linux can indeed be a secure platform. Due to the nature of Linux development, security fixes often come out much faster than they do on commercial operating systems, making Linux an ideal platform when security is a requirement.

Acknowledgements
Information here is collected from many sources. Thanks to the following that either indirectly or directly have contributed: following who either indirectly or directly have contributed: Rob Riggs [email protected] S. Coffin [email protected] Viktor Przebinda [email protected] Roelof Osinga [email protected] Kyle Hasselbacher [email protected]
David Jones (20.01.00) Page 477

85321, Systems Administration

Chapter 17: Security

David S. Jackson [email protected] Todd G. Ruskell [email protected] Rogier Wolff [email protected] Antonomasia [email protected] Nic Bellamy [email protected] Eric Hanchrow [email protected] Robert J. [email protected] Ulrich Alpers [email protected] David Noha [email protected] The following have translated this HOWTO into various other languages! A special thank you to all of them for help spreading the linux word... Polish: Ziemek Borowski [email protected] Japanese: FUJIWARA Teruyoshi [email protected] Indonesian: Tedi Heriyanto [email protected]

David Jones (20.01.00)

Page 478

85321, Systems Administration

Chapter 17: Security

David Jones (20.01.00)

Page 479