$9

Using
Unix
Get started with the command line
by Casimir Saternos
Unix Basics

©2008 Casimir Saternos

Every effort was made to provide accurate information in this document.

However, neither the author nor Topfunky Corporation shall have any liabil-
ity for any errors in the code or descriptions presented in this book.

“Rails” and “Ruby on Rails” are trademarks of David Heinemeier Hansson.

This document is available for US$9 at PeepCode.com (http://peepcode.com).

Group discounts and site licenses can also be purchased by sending email
to [email protected].

2
 contents 30 Tow the (Command) Line 52 Somebody’s Watching Me…
56 Monitoring File Space
8 About UNIX 33 Organization
56 Monitoring a Process
8 History 33 Moving in Stereo
57 Scripting
9 Philosophy 33 It’s in the Files
57 The Job That Ate My Brain
9 Do One Thing Well 36 Find Your Way Back
58 Debugging
9 Output is Input
39 Concatenation
10 Everything is a File 61 Appendix: Online Resources
39 Here, There and Everywhere
10 Help! 62 Appendix: Info Script
39 Sequencing
13 Navigation 39 Redirection
13 Starting Here, Starting Now 40 Appending
41 Pipes
14 Where Am I?
41 Filters
15 Path Descriptions
46 I Know the Difference
17 Where Am I Going?
47 Time After Time
18 Who Am I? 48 World on a String
20 Power Tip: SSH Keys
50 Administration
23 Modification 50 Installation
23 Editing Files 50 Unix/Linux Installation
25 Permissions 51 Starting Up / Shutting Down
26 Aliases and Inodes 52 Users/Group Administration
52 System Monitoring
30 Customization 52 Backup/Recovery
about this book
Using UNIX can be intimidating! If you are familiar with Windows or
the graphical elements of Mac OS X, you may be overwhelmed when
presented with a blank terminal screen, waiting for a command.

But don’t panic! You only need a dozen short commands for most of
what you’ll need to do as a web developer. As you become comfort-
able, you’ll learn to customize your environment for greater efficiency.
This book is designed to provide the information to quickly and effec-
tively address your daily challenges and is extensive enough to give
experienced developers insights that will increase their effectiveness
and productivity.

note Some developers have blogged about their most frequently

used commands. They can be found by searching
Google for command history meme (http://www.google.com/
search?q=command+history+meme).

The purpose of this book is to give you insights into how to develop,
install, monitor, and maintain web applications on UNIX servers. We’ll
assume that either:

• You have access to a UNIX server (a shared host, a virtual host, or a

dedicated server), or,
• You’re using Mac OS X (which uses BSD UNIX)
The intended audience for this book is web developers who create,
deploy, maintain, and troubleshoot web applications that run on UNIX
servers. Developers planning to deploy to a Linux-based virtual server
are faced with a challenge if the majority of their development was
completed on a Windows or Mac OS.

Sometimes web developers naively think they can avoid learning

about UNIX details that can be left to system administrators, but this
4
is not a realistic ideal. Many Rails developers deploy their web appli-
cations with Capistrano. Although it utilizes Ruby-based configura-
tion files, a quick glance at Getting Started with Capistrano (http://www.
capify.org/getting-started/basics) reveals that the often used run function is
simply a means to execute system commands.

You should be comfortable working from a command line. You do

not need to know advanced shell scripting techniques or anything
like that (though it helps, if you want to start writing complex
Capistrano recipes), but you should be able to navigate directories
and execute commands from the command-line. Capistrano
has no GUI interface; it is entirely command-line driven.

So to be effective with Capistrano, you need to be conversant with the

commands available on the underlying operating system (and better
yet, bash shell scripting).

note We’ll use the terms terminal, command line, command

prompt, and shell to refer to the application used to enter
UNIX commands.

With web development in mind, there are a number of subjects tra-

ditionally covered in books on UNIX which are not addressed here.
System administration tasks are limited to the types that develop-
ers are most likely to address and will provide enough information for
you to communicate effectively with system administrators. There is
no coverage of kernel development, file system features, or graphical
user interface environments except when a subject is related to the
work of a web application developer. The aim is to filter out the noise
and give you immediate access to the features that will help you in
your development tasks.

Even if you are not using UNIX on a daily basis, it is worth noting that
UNIX has exerted a considerable influence on the design of other
5
technologies. Mac OS X, Tivo, routers, and hand held devices run
Linux under the hood. In addition, many of the utilities have corol-
laries in classes in the Ruby programming language. See the chart
below for some examples of features of Ruby that are based upon
UNIX utilities.

Command FileUtils String Array

cd X
pwd X
mkdir X
ln X
cp X
mv X
rm X
chmod X
chown X
touch X
grep X
tr X
uniq X
sort X

Numerous other examples might be mentioned including regu-

lar expressions and the date, time, and find classes. The bottom
line is that becoming better acquainted with UNIX functionality will
make you a more effective developer even when you are not working
directly on a UNIX system. It introduces you to a way of thinking—and
a programming culture—that continues to exert a significant influence
on software design decisions today.

There are a number of UNIX variants, branches and distributions

6
(called distros) of UNIX operating systems. Each of these have spe-
cific features and unique implementations that differentiate them
from their siblings. However, there is a rather large body of common
functionality which allows most users to use a consistent set of com-
mands on each system. This book will focus on functionality that is
available to the majority of those in popular use.

The commands that are introduced in this book will be described, but
not in a comprehensive manner. The book can be read from begin-
ning to end to provide a relatively comprehensive overview of UNIX
that focuses on the interests of web developers. You can also read
individual sections to get an introduction to the resources available
in a particular problem domain and specific examples of how to use
UNIX effectively for a particular task.

7
About UNIX
chapter 1

History
In the mid 1960’s, a group of companies set out to build a comput-
ing utility service that could be used by many people at once like the
phone or electric utilities. MULTICS (http://en.wikipedia.org/wiki/Multics) never
became as popular as originally hoped, but some of the engineers
who worked on it went on to develop a more focused operating sys-
tem called UNIX.

UNIX was originally developed in 1969 at Bell Labs. Throughout the

1970’s UNIX was used and improved upon, primarily by scientists
and technicians in research and educational settings. By the late
70’s UNIX began to be adopted in business settings by students who
had done undergraduate and graduate work on UNIX. During the
next decade a number of different versions of UNIX were in use and
standards related to UNIX were developed. Linus Torvalds introduced
Linux in the early 1990’s as a rewrite of MINIX (http://en.wikipedia.org/wiki/
MINIX) (a minimal UNIX operating system).

It’s important to realize that UNIX is an operating system that was

written by programmers for programmers. Programmers tend to think
of writing programs in their language of choice to solve problems.
However, it is possible to solve complex problems using nothing but
built-in UNIX functionality. You might think of UNIX as a component
architecture in this context. With a proper understanding of the pro-
grams available (and how they can be used together) you will often
find yourself quickly solving problems through a script or series of
commands without the overhead of having to resort to coding an
entire independent program.

8
Philosophy
The UNIX philosophy was never formalized, but here is an attempt to
summarize the fundamental design principles that guided the devel-
opers of the operating system.

Do One Thing Well

Throughout this book you will encounter UNIX commands and utili-
ties. Each of these is designed to perform a single task well. New
programs were created when there was concern about extending the
functionality of a program in a way that would confuse the original
intent of a program. Each command has a short, descriptive, lower-
case name that reflects its central intent.

note Most UNIX commands are only a few characters long. You
may want to shorten commonly used commands even further
with aliases, which we’ll talk about later.

Output is Input
Very few UNIX utilities require interactive input, and their output is in
a form that is suitable for use by another program: plain text! This
makes it easy to use several programs together, even if they aren’t
aware of the other program. Programs can be chained together in a
sequence, and some programs have a primary usage as filters for
other programs. Each filter in a series of commands serves one small
purpose in modifying a stream of text before it is transformed into its
final output.

note A simple example is wc, a program that reports on the

number of characters, words, and lines in any string of text.
9
Everything is a File
Files can be classified as ordinary files, directories, or special files
in UNIX. However, everything that the OS is concerned with is rep-
resented as a file. This means that diverse resources such as docu-
ments and directories as well as hard-drives, CD Drives, DVD Drives,
keyboards, printers, monitors, and terminals are all accessible through
a common method of communication.

This results in some nonintuitive uses of commands. For instance,

according to the man pages the lsof command lists open files.
However, you can use the utility to determine the program listening
on port 80 (lsof -i :80). Network resources are also represented in
files in this context, which is a source of confusion to the uninitiated.

Help!
Since this book is not comprehensive, you should be aware of how
you can expand your knowledge on your own in a particular area of
interest. There are a huge number of utilities, scripts and programs
available on most standard UNIX installations. You will most likely find
that there are several thousand programs available in yours—some of
which are so complex that entire books are dedicated to them.

The man and info commands are the standard utilities to use to learn
more about each command you encounter. The man command can
be called with a program as the argument, and the program’s manual
page will be displayed. If you are not sure of what program you need,
the -k option will be of assistance. For instance, type the following to
get information on programs related to the Apache web server:

% man -k apache
ab(8) - Apache HTTP server benchmarking tool
apachectl(8) - Apache HTTP Server Control Interface
apxs(8) - APache eXtenSion tool
10
 httpd(8) - Apache Hypertext Transfer Protocol Server

Once a given utility has been identified, you can simply type man
followed by the utility name to see the manual page. Press space to
scroll through the page and q to quit. And in typical recursive fashion,
you can man man to find what other man options are available.

% man ab

NAME
ab - Apache HTTP server benchmarking tool

SYNOPSIS
ab [ -A auth-username:password ] ...

Command Description
man Display the manual page.
Press space bar to scroll
through and q to quit
man -k Search the whatis data-
base for commands that
match
man -f Equivalent to whatis, which
provides a short description
of a system command
info Display an info page (more
comprehensive than man)
<command> -h Display the help screen
type -t Identify a command as an
alias, keyword, function,
builtin, or file

11
12
Navigation
chapter 2
You’ll learn best if you try to follow along with these examples or the
accompanying screencast.

Starting Here, Starting Now

If you’re using Mac OS X, you can follow along with most of this book the prompt
using the Terminal application on your local machine. Otherwise,
you’ll need to connect to a UNIX server. The command line prompt can be custom-
ized. You can put almost any information into it,
including:
×-+

  • Your username: csaternos

• The name of the host your commands are
running on
• Your current directory: ~/tmp
• Contextual information, such as the name of
the current Git repository branch
We’ll use a % to signify the place where com-
mands are issued.
Fig. A Command line prompt

There are a number of ways that you might access a server. Although
there are a variety of graphical desktop options available when log-
ging at the console or through VNC (remote screen sharing), the focus
of this book is going to be command line access. Much of the power
of the UNIX operating system can only be realized when accessing
the system from the command line.

The telnet and rsh (remote shell) applications are old, standard,
13
insecure means for accessing systems. It is more common (and
secure) to use ssh today.

On Mac OS X, go to Applications • Utilities • Terminal. On Windows,

PuTTY (http://tinyurl.com/2r4w) is a popular SSH client that can be used
to connect to a UNIX server.

To connect to a remote server, you’ll need the following:

• Your username, such as csaternos

• Your password
• The domain name of your server, such as peepcode.com.
To connect from the command line, run this command with your
information:

% ssh [email protected]

You will be prompted for your password and can now issue com-
mands on the remote server.

note Later we’ll talk about how to login automatically without a

password using SSH keys.

Where Am I?
Logging in locally or via SSH will drop you into your home directory.
This is the directory where you can store files and create directories.

The pwd command (print working directory) can be used to print your
current location in the filesystem. The cd command can then be run
to navigate to a new directory. Within a directory or set of directories,
14
you can view the contents using the list (ls) command.

Path Descriptions
Many commands take a filename as an argument. The argument
can be an absolute path, a relative path, or a special character that
designates a directory. Here are a few:

Path Description
~/ Your home directory
/Users/topfunky An absolute path. Starts
with a slash and specifies
all the intervening direc-
tories needed to get to a
location
./bin A relative path. The two
periods and a slash signify
one directory up from your
current location. These can
be added an several times
to navigate up several
directories.
- The last directory you were
in. This might be in a com-
pletely different location
and is useful for returning
there (with cd) or copying
files.

The ls command includes a number of options that allow you to

filter, sort, and format file listings as well as to specify relevant infor-
mation about the files listed. By combining the results of an ls
command with other utilities (such as awk, grep, sort, and uniq) a
15
huge number of options become available. Although it’s a well-known
command, many people are not familiar with the wide range of uses
available to ls.

The use of an asterisk as a wildcard or globbing pattern is a common

idiom. Although glob is not a command, you can man glob for more
information. If you need to include a wildcard name in a command,
you can indicate that the wildcard is to be treated as a literal by the
shell by enclosing the name in quotes.

Wildcard Description
? Represents any single
character to be matched
* Represents any number of
characters to be matched
[] Specifies a range matched
{} Contains a list of comma-
separated values, any of
which will be matched
[!] Excludes a range matched
Used as an escape char-
acter to treat a special
character as a literal

The tool time is noteworthy when trying to track down what file in a
group has changed recently. To sort the results of time, use the -t
option. If you are more concerned about the size of the files, include
the -S option. Piping the results to more or less will allow you to page
through a large list of results. Grab the first line in a file piping to head
-1, or the last line using tail -1.

If you want to use the the listing of files as string literal input to a sub-
sequent program, you can use the -Q option to enclose each name in
16
quotes.

If you wanted to remove a set of files that you have listed, you could
use xargs command (rather than directly piping to the rm com-
mand):

% ls \*/\*diffs.txt | xargs rm

Command Description
cd Change directories
ls List files
xargs Build and execute com-
mand lines from argu-
ments in standard input

Where Am I Going?
Once you have logged on to a server, you will want to get familiar
with what resources are available and begin making customiza-
tions to meet your particular needs. Typically, you will immediately
check your that your environmental variables (e.g. PATH) are set as
intended, and perhaps validate the system specifications.

In the bash shell, a file named .bash_profile within the user’s home
directory contains environmental settings specific to the user. You
might want to adjust the settings in this file. In addition, there are
times that you might source this file (execute the script in such a
way that the variable settings become present in the current session).
Scripts scheduled through cron often have this requirement. The way
to source your bash profile is as follows:

% . .bash_profile
17
 tip See dotfiles.org (http://dotfiles.org) for profile samples from other
developers.

You can use the env command to see the settings of variables within
your environment. You can use export to make new variable settings
available to the environment.

Command Description
env Outputs the variables that
are set in the environment
export Make variable settings
available to the current
environment
alias Display or set substitutions
for commands or groups of
commands
who Display information about
users that are logged in
uname Output information about
the system (kernel release,
hardware info, processor,
etc)
hostname Output the name that
identifies the server

Who Am I?
As in the rest of life, it’s important to know who you are.

18
Because UNIX supports multiple users and each user has specific
privileges, it is important to understand who you are logged in as.
Problems can occur when a file is created by one user and a later
attempt is made to access the file by another user.

Most installations of UNIX start out with several user accounts and
groups. Here are a few that may be present on your system:

User Description
root The administrative super-
user account. The root user
can perform any com-
mand on the system. If an
attacker gained access to
the root account, he could
take over the rest of the
system.
nobody An unprivileged account
used by applications that
want to restrict access as
much as possible.
daemon A faceless account used to
run background tasks.

Rather than logging on with root privileges, a user can be granted

access through the sudo command to perform privileged operations.

In this example, I’ll try to find the file sizes of all the directories on the
current machine. Some are protected, so I’ll have to rerun the com-
mand with sudo in order to have access to all the files.

% du -hs /*
du: `/usr/share/ssl/CA’: Permission denied

19
 % sudo du -hs /*
Password:

3.6M /bin
2.7M /boot
568K /dev
4.5M /etc
772M /home
1.3G /usr

Command Description
whoami Identify the user you are
currently logged in as
passwd Change your password
su Switch user
sudo Execute a command as
another user. Frequently
used to allow super user
(root) access
visudo Editor for the sudoers con-
figurations file which limits
access available through
sudo

Power Tip: SSH Keys

It’s tedious to type in a password every time you need to connect to
a server.

If you are interested in being able to log in without being prompted

for a user name and password, you need to set up a private/pub-
lic key pair using the key generation utility. Rails developers plan-
20
ning to use Capistrano are required to use ssh to make connections.
Because of the requirement of using the same password on all serv-
ers when using regular logins, it is advisable to access your servers
using a public key using ssh-keygen.

The following steps can be used to generate a key pair, transfer the
public key to a server, and log in (or copy files via scp) without a
password. In this example, the destination server name is destina-
tionserver and the OS username is webuser.

1. Generate the key on the client.

% ssh-keygen -q -f ~/.ssh/id_rsa -t rsa -C “Demo of Key Generation”

2. Transfer the Public file (id_rsa.pub) to the server in a file named

authorized keys in the .ssh directory within the user’s home directory.

% scp ~/.ssh/id_rsa.pub webuser@destinationserver:.ssh/authorized_keys

3. Login and transfer are now available from the client to the server
without a password.

% ssh webuser@destinationserver
% scp source_file.txt webuser@destinationserver:destination_file.txt

It is generally a bad idea to log on directly as root for normal main-

tenance. If multiple users log on as root, there is no way of auditing
the activities that have occurred on the server. The sudo command
(and visudo command which allows an administrator to maintain the
sudoers configuration file) can be used to provide auditable access
to actions and resources reserved for superusers.

21
Command Description
ssh Secure shell remote login
ssh-keygen SSH key generation and
management

22
Modification
chapter 3

Editing Files
Because everything in UNIX is a file, it is crucial to be conversant with
the options available for viewing, editing, and manipulating files. The
vi editor is a common, powerful interactive text editor. You will need to
invest a bit of time to be somewhat conversant with it, but is well worth
the time. It also serves as the editor for a number of special editors
that are designed to safely modify certain system configuration files.

Special Editor Description

crontab -e Edit the file used to specify
special processes run by
cron
visudo Editor for the sudoers
file that allows controlled
access to privileged opera-
tions
vipw Editor for the password file
vigr Editor for the group file

Although interactive editors are irreplaceable, there are many tasks

that can be completed more easily and accurately using stream
editors, commands, or scripts. The contents of two files can be con-
catenated using the cat command, but a more common usage is
to pass a single filename as a parameter and view the results on the
screen. (The lesser-known tac command functions the same way but
processes files in reverse). When viewing the contents of a larger file
using cat and you are interested in viewing the contents a screen at
a time, the results can be piped to more or less. Amusingly enough,
23
less is more and, well more. If you are interested in only viewing the
beginning or end of a file, head or tail can be used respectively. A
common usage of the tail command is to view the end of a log as
output is added to it.

For example:

% tail -f production.log

Press CTRL-C to escape.

The sed stream editor can be used to do search and replace opera-
tions in files. When more complex stream-oriented modification of
files is needed you will likely want to switch over to ruby or perl. Finally
redirection to a file using the > (create new) or >> (append) operators
is a way to write the results of a command to a new or existing file.

Command Description
cat When passed a single file
as an argument, displays
the contents to standard
out
tac Concatenate/print a file in
reverse
more Filter for viewing text one
screen at a time
less Functions like more but
allows backwards scrolling
and other features
head Output the first part of a
file

24
Command Description
tail Output the last part of a
file
vi An editor standard to most
UNIX installations
emacs An extensible customizable
text editor that includes a
Lisp interpreter

Permissions
Many problems encountered by new UNIX users are related to setting
file ownership and file permissions. Permissions can apply to users,
groups or others.

The read, write and execute bits are the standard permissions and
there are also some special modes that can be set. There is an octal
notation that is used to specify file permissions summarized the chart
below.

Permission Description Digit

-- No permissions 0
-x Execute 1
w- Write 2
-wx Write and Execute 3
r- Read 4
r-x Read and Execute 5
rw- Read and Write 6
rwx Read/Write/Execute 7

25
You can change both the owner and the group of a file at one time:

% chown oracle:dba x.txt

The octal notation or the letters can be used to specify permissions.

The command to change group, owner, and all permissions is listed
below using letter abbreviations and octal format.

% chmod goa+rwx x.txt

% chmod 777 test.txt

Command Description
chown Change Owner of files/
directories
chmod Change Modification of
files/directories
chgrp Change Group of files/
directories
umask Set the default file system
mode for newly created
files and directories
alias Display or set an alternate
name for a command

Aliases and Inodes

There are several different names that can be used to specify com-
mands and files in UNIX. As mentioned earlier, the ln command can
be used to either specify another name for a file or provide a pointer
to a file. An alias can be used to explicitly specify a shortcut (includ-
ing a parameter or even an entire command sequence) for a com-
26
mon activity. Besides these mechanisms that allow you to create new
names, the file names that we ordinarily act upon are really pointers
to a numerical designations for each file, known as inodes.

aliases
The @alias@ command with no arguments displays the aliases in use
in the shell. Aliases can be used to help users who are familiar with
commands on other operating systems.

% alias dir=’ls -l’

They can also be used to correctly interpret typos (e.g. aliasing ls-l
which is missing the space between the command and option to
ls -l). Commands such as cp, mv and rm can be aliased to prompt
before overwriting or removing files.

An alias can be removed by using the unalias command. Aliases

are expanded by default in interactive shells (see the shopt com-
mand expand_aliases option for details). This can result in unex-
pected behavior when they are used in scripts.

inodes
You can use ls -i to print the index number, or inode of each file.
The stat command will display the inode number as well as its
status and attributes. An inode contains all of the file’s properties
(permissions, ownership, type of file) as well as the addresses for the
data blocks containing the file’s content. Several file names can link
to the same inode. The rm command removes the link that points to
its inode, but does not remove the file contents (so the inode has no
name linking to it at all). If another process still has the file open, you
can even restore a deleted file.

% echo “This file will be removed...” > file_to_remove.txt

27
 % echo “But we will get it back.” >> file_to_remove.txt

Now let’s force a process to hold on to the file while we deleted it. We
open it with more but only read a line at a time

% more -1 file_to_remove.txt

Type Ctrl-Z while more is still running. This will force the process into
the background. The following line is initially displayed and also will
be seen if you type jobs.

[1]+ Stopped more -1 file_to_remove.txt

Now let’s remove the file

% rm file_to_remove.txt

Try to ls or view file. It does not appear to be available. However, the

file is still open by the more command as can be seen using the lsof
command.

% lsof | grep file_to_remove

% more 26133 csaternos 3r REG 8,5 54 4823091 /home/cs/file_to_remove.txt (deleted)

The output from the command provides the information necessary to

restore the file by copying it out of /proc. The contents of the original
file are designated as follows:

/proc/<process id from column 2>/fd/<number embedded in column 4>

28
Based upon the output displayed above, we could run a command to
restore the file and view the contents:

% cp /proc/26133/fd/3 restored.txt
% cat restored.txt

Command Description
alias View or create aliases
unalias Remove aliases
lsof List open files
shopt View or modify shell
options

29
Customization
chapter 4

Tow the (Command) Line

Once you are logged onto the server, you will want to collect some
basic information about what is available to you. The shell is an
interface of sorts to the underlying operating system. It dictates what
built-in commands are available to control the OS kernel.

There are a number of different ones available. The Bourne Again

shell (bash) is the default shell on the Mac OS X and most Linux
systems. It is available on most UNIX systems and will be the shell
assumed for the rest of the book.

Once you become familiar with the basics, you may want to look at
the unique features of other shells:

• zsh (http://www.zsh.org)
• fish (http://fishshell.org)
• tcsh (http://www.tcsh.org)
Shortcuts are a key factor in improving your accuracy and produc-
tivity in any environment. The table below lists some of the popular
ones. Others are available online at the Nuby on Rails blog (http://
nubyonrails.com/articles/useful-shell-shortcuts).

30
Shortcut Key Description
Ctrl-R Search your bash history.
Type the first few letters
after hitting Ctrl-R and
press enter when the com-
mand is selected
Ctrl-T Swap the last two charac-
ters you typed
Ctrl-C Cancel the currently run-
ning command
Ctrl-D Log out of the current ter-
minal
Ctrl-Z Stop a running Job (see bg
and fg for more informa-
tion )
!! Repeat the last command
!$ Recall the previous argu-
ment
<arrow-up>/<arrow-down> Scroll through Command
History
<TAB> Complete a command or
name based upon entries
available in a given context

Being able to repeat the previous argument allows for convenience

and accuracy (fewer key strokes) as well as safety when executing
commands.

Command Description
history Print the command history

31
Command Description
whatis Short description of a com-
mand
clear Clear the screen
stty Change or display terminal
settings

32
Organization
chapter 5

Moving in Stereo
Files can be moved around the file system using mv and copied using
cp. Moving files between computers involves the use of other com-
mands or protocols (ftp, scp).

Moving large files around can be a challenge, but numerous options

are available. Files can be compressed using a number of different
utilities including tar, zip and cpio. Large files can be broken into
more manageable bits using the split command. However, for large
files that need to be kept in sync across servers the rsync is the most
efficient means to minimize the amount of data that needs to be
transferred.

It’s in the Files

Shortcut Key Description

touch Change the file time or
create a file if it does not
exist
ln Create a link (short-cut) to
another file
sed A stream editor

If you have a script where a command relies upon the existence of a

file, you can either test for the file’s existence and take an appropriate
action or simply touch the file in advance to ensure that it exists.

33
Links are analogous to Windows shortcuts. Symbolic links allow a
given symlink or soft link with one name to refer to a file with another.
Some operations (like an rm) operate on the link itself but most
operations affect the targeted file. Symbolic links are created using
the -s option:

% ln -s /<target> <my_link_name>

Hard links function as a different name of an existing file. Hard links

are generally not allowed for directories (unless a special option is
specified for a super user). You can, however make symbolic links to
directories. You might want to create a link for a directory if you have
an additional disk added to your system that is available on a differ-
ent file system. However, it can be a bit confusing when working with
links to directories because of the fact that some operations work on
the destination file and some work on the link. Consider the following:

Start with an empty directory

% mkdir testing
% cd testing
% ls -lth
total 0

Create a Symbolic link to another directory:

% ln -s /work
% ls -lth
total 0
lrwxrwxrwx 1 csaternos users 5 2008-03-20 09:34 work -> /work

Navigate to that directory

34
 % cd work

Now let’s look at the parent directory—Oh wow—it’s the parent of the
destination!

% ls ..
bin boot dev etc home lib mnt opt proc root sbin sys tmp usr var work

Hey let’s go up a directory…

% cd ..

We are back in the directory where originally created the link

% ls
work

There are numerous interactive editors available for UNIX. The classic
vi editor is available on nearly all platforms and emacs has become
a favorite among rails developers. One early editor is named ed, and
a stream oriented variant of this editor is named sed. The sed pro-
gram can be used to do search and replace operations in files.

% sed s/black/white/ original.txt > modified.txt

If the tasks become more sophisticated, perl and ruby can be used to
your advantage. In particular, one-liners may be created that require
a simple invocation of the perl or ruby interpreter along with argu-
ments that include a few commands and files to process.

35
Find Your Way Back
The ls command is sufficient when viewing the contents of a single
directory or a few directories withing a given hierarchy. However, it is
not useful for identifying files that are in a number of different loca-
tions. The find utility can be used to search for files in a more sophis-
ticated manner. One simple example to get you started:

% find . -name *.rb -print

The dot indicates to search within the current working directory. The
*.rb indicates that any file name ending with the .rb extension is an
eligible target. Other file attributes (e.g. size, permissions, type) can
be used; for instance, this command shows all files within the current
working directory that have been modified in the past 5 minutes:

% find . -mmin -5

The find command is not limited to simply listing file names. find can
execute commands directly, and you can also pipe results to xargs:

% find . -name *pc.rb | xargs ls -l

This opens up all kinds of possibilities—executing every file found,

removing files of a certain type or age, etc. At some point, it becomes
simpler (or perhaps clearer) to write a script within ruby to accomplish
the task:

#!/bin/env ruby
require ‘find’
Find.find(‘.’)do |x|
# Process the file here...
end
36
The find command is very flexible and comprehensive in its ability to
search for files, but it can be slow because of its design. The locate
command relies upon the existence of a database of files. Typically
a system is set up to run a process on a regular basis to keep the
database up to date. The locate command is quicker than find, but
is only as accurate as the database it is accessing. In the following
example, the locate database needs to be updated before it can be
used to find a file.

% locate test.txt
warning: locate: warning: database
/var/lib/slocate/slocate.db’ is more than 8 days old

% locate -u .
% locate test.txt
/root/Desktop/download/Python-2.5.2/Lib/test/test_doctest.txt
/root/test.txt
/usr/local/lib/python2.5/test/test_doctest.txt

The find and locate commands are great general purpose utilities,
but a rather common need is to determine the location of a program
or determine the version of the program in use. The whereis and
which commands are targeted for this usage.

Command Description
find Search for files within a
given directory hierarchy
locate Faster than find—but uses
cached information in a
database

37
Command Description
whereis Display the location of the
binary, source, and man
page
which Show the full path to
commands—useful for
determining the version of
the command in use

38
Concatenation
chapter 6

Here, There and Everywhere

Many programmers know how to execute a few simple commands
on the UNIX OS. The real power of UNIX becomes more evident when
commands are combined in a creative fashion. The next several ses-
sions introduce the concepts related to combining functionality of
several commands.

Sequencing
In simplest form, you can execute one command per line. However,
you can execute several commands in sequence by separating them
by semicolons. Note that the input and output of each command is
independent of the others in the series.

Redirection
Before UNIX, programs generally had to explicitly connect to the
appropriate input and output data. Standard Input (STDIN) is data
that serves as input to a program. It is expected to come from the
text terminal that started the program. However, you can redirect
input instead. Standard Output (STDOUT) is the output from a pro-
gram. The output by default is directed to the text terminal that
started the program, but this can be redirected as well.

A basic example of redirecting STDIN and STDOUT follows:

% echo “GoodByeWorld” > test.txt # Redirecting STDOUT to a text file

39
 % read x # Reading from STDIN
HelloWorld

% echo $x # Displaying to STDOUT

HelloWorld

% read x < test.txt # Redirecting to STDIN

% echo $x # Displaying to STDOUT. See, it worked!

GoodByeWorld

Standard Error (STDRR) consists of error messages produced by a

program. Its output, like that of STDOUT, is directed to the text termi-
nal that started the program. However, it is a separate stream which
is useful to keep errors and diagnostic messages separate from the
actual output of a program.

To redirect to standard error:

% echo “An error occured” >&2

To redirect standard error to a script:

% my_script.sh 2> errors.log

Both STDERR and STDOUT can be redirected to a file using:

% my_script.sh > output.txt 2>&1

Appending
To append to the end of a file (when using STDERR or STDOUT),
40
change the > to >>. For example

% echo this > test.txt

% echo that >> test.txt
% cat test.txt
this
that

Pipes
The output of one program can serve as the input to another one.
This is one area where UNIX really shines. Many UNIX users never
make the connection that the many small utilities can be combined
to very powerful effect. Pipes serve as the connectors between utilities
that are combined to perform tasks.

Redirect data to a file and write it on STDOUT at the same time

% echo “It’s tee for two!” | tee result.txt

It’s tee for two!

% cat result.txt
It’s tee for two!

The pipe redirects STDOUT to STDIN of the next process. The tee
command reads from STDIN and redirects to STDOUT and the file
result.txt at the same time

Filters

41
sorting/ordering
Many commands include options that allow you to sort their output
(such as ls discussed earlier). If you need to order the results in a
command in a way that is not possible for the command in question,
you can use the sort command. The sort command allows sorting
that varies with space used, type of element (numeric, alphabetic etc),
and location of the element that serves as a key for sorting.

The sort command is often used to filter input that is passed to the
uniq command. Uniq removes duplicates from the list of lines passed
to it. If the lines have not been previously ordered by sort, the out-
put of the uniq command will only eliminate duplicate entries that
appear sequentially in the list of lines.

The sort command includes an option that allows you reverse the
output of an other command. To sort on the 6th field first use -t,
switch to tell sort that the delimiter is a comma and -k to tell sort
which field to sort on:

% sort -t, -k6 file

Descending on the same field:

% sort -t, -k6 -r file

But there are also several other commands that can be used to
reverse output in various ways. To illustrate, lets start by producing a
list of zero-padded numbers between one and eleven.

% seq -w 11

We can reverse the list by piping to tac (which works like cat in
42
reverse):

% seq -w 11 | tac

We can reverse each element in the list using the rev command:

% seq -w 11 | rev

limiting
There are a number of different options available to limit the output
of files. The grep utility and awk programming language (that is com-
monly used to execute one liners) have a broad range of functionality
that are well beyond the scope of this book. A few minimal examples
will suffice to get you started.

In general, use grep to decide which lines you want displayed in a

group of processed lines. Regular expressions can be specified to
identify a wide range of strings. The following example allows you
to identify every line that contains the word split in files with a ruby
extension.

% grep split *.rb

The name of the file in question will be displayed before each line if
there are multiple entries. If you wanted to list every file that did not
contain the word split, you would use the -v option.

% grep -v split *.rb

The following example uses a regular expression that lists lines that
include 2 or 9.
43
 % seq -w 1000 1010 | nl | grep -E ‘2|9’

The somewhat cryptically named awk language is derived from the

final initials of its designers: Alfred V. Aho, Peter J. Weinberger, and
Brian W. Kernighan. The awk utility can also be used to limit which
lines are displayed.

% cat *.rb |awk ‘/split/{print}’

But awk is particularly useful for quickly breaking up a line into a a

series of tokens based upon a delimiter (by default a space). Each
token is identified by a dollar sign followed by the number repre-
senting the order where it appears (starting with one). The following
example lists all of the shell scripts in the directory and displays their
timestamp in YYYY-MM-DD format.

% ls -l --time-style=long-iso |awk ‘/.sh/ {print $6}’

If you want to use a different delimiter, use the -F option and specify
the delimiter name.

% echo “one,two,three” | awk -F, ‘{print $2}’

enhancing
Enhancing might not be the best way to describe this type of filtering,
but adding line numbers to output is a common requirement.

% cat environment.rb | nl

44
grouping
It is common to want to redirect the output of a number of a group of
commands. Rather than individually specifying the redirection:

% ls >> output.txt
% df >> output.txt
% who >> output.txt

the commands can be rewritten enclosed in parenthesis.

% (ls
df
who) >> output.txt

The uniq utility with the -c option provides a count of unique entries.
This is somewhat analogous to the use of a GROUP BY in SQL. The
following example displays each file owner within the directory fol-
lowed by the number of files that they own:

% ls -l | awk ‘{if ($1 != “total”) print $3}’ | sort | uniq -c

Command Description
grep Print lines that match a
specified pattern
awk Special purpose pattern
matching and data format-
ting language
sort Order lines based on
specified criteria
uniq Remove any duplicates
from input
45
Command Description
nl Number lines
tac A version of cat that reads
through a file from the end
to the beginning
rev Reverses inputted lines

I Know the Difference

The original diff utility used for file comparisons was written in the
early 1970’s. Since then, numerous enhancements and variations
have been created. The diff utility displays the differences between
two files on a line by line basis. The output of diff is in a format that
can then be used by the patch utility to apply the changes to a file.
The sdiff utility produces a side by side comparison between two
files. The comm utility produces three columns of output that show dif-
ferences side by side—but requires the files to be sorted.

If you simply want to see the side-by-side output of two files to make
comparisons for yourself, the paste utility can be used. The join
command can be used to display the lines with a common key field
between two files. Although this is useful for simple comparisons,
comparisons using key fields are better handled in an RDBMS.

If you are required to reconcile the differences between three files,

the diff3 utility is available. There are a number of GUI programs
(kdiff3) that can be used to display differences with color-coding
and other visual cues. Such programs also include options for com-
paring entire directories to identify differences.

46
Command Description
Diff Display the differences
between two files
patch Use the output of diff and
apply it to a file
cmp Compare two files
join Join lines in two files using
a key field
paste Combine two files side by
side
comm Compare two sorted files
side by side

Time After Time

With no parameters, date returns the current date. You can also
manipulate date strings passed in as parameters or in files. For
example:

% date -d 19700201 +%m/%d/%Y

This example might not work on all implementations (I noticed that

it does not work using MKS Toolkit which provides UNIX emulation
on Windows). See the implementation below using substrings for an
alternative. To determine the length of time a command or script
takes to execute, you might be inclined to do something like:

% date;<my long running command>;date

However, you would be better off using time:

47
 % time <my long running command>

This command returns timing statistics including the elapsed time,

user CPU time and system CPU time.

Command Description
date Return the system date or
manipulate date strings
time Measure the resource
usage and elapsed execu-
tion time of a command
cal Display a calendar

World on a String
The bash shell provides for string manipulation—but in a rather incon-
sistent manner. Some commands are based on parameter substitu-
tion while others require the expr command.

The following function receives an input of a date string in MMDDYYYY

format. It returns the date in MM/DD/YYYY format.

% function slash_date {
d=$1
echo ${d:0:2}/${d:2:2}/${d:4:4}
}

The second line in the function could also be written as follows:

% echo `expr substr $d 1 2`/`expr substr $d 3 2`/`expr substr $d 5 4`

48
The length of a string can be outputted as follows:

% expr length “This is twenty chars”

A string can be converted from upper to lower case using the tr

command.

% echo “Hello” | tr “[:upper:]” “[:lower:]”

A default value can be assigned to a string as follows:

% echo ${username-`whoami`}

Command Description
expr Evaluate an expression
tr Translate or delete charac-
ters in a string

49
Administration
chapter 7
As mentioned earlier, the focus of this books in not on system admin-
istration, and so this section is intentionally brief. It is intended to
serve as a point of reference for communication with system adminis-
trators or as a stepping stone to further investigation.

There are a myriad of installation options that are dependent upon

the particular UNIX variant in use, available hardware, and network
configuration. This section will provide a brief, high-level overview of
these concerns. The remainder of the book will assume that this work
has already been completed—exactly the situation that you will find
yourself in when logging into your Internet Service Provider or access-
ing a system within a corporation’s internal network.

Installation
Tools such as apt-get, yum, and port can compile and install soft-
ware easily. See the accompanying screencast for an example.

Unix/Linux Installation
If you are installing an operating system for the first time, you will
probably perform the installation in an interactive manner. If you are
going to have to repeat the installation, you should probably consider
automating the installation process.

Part of the initial configuration involves the modification of certain

system files to set kernel parameters, mount points (/etc/fstab),
and other system settings.

50
Besides the base operating system, you will have to install other
software (web servers, databases, other language environments,
and packages). You will need to become familiar with the installation
utilities and package manager (rpm) associated with the particular
system in use. You may need to build from source using a series of
commands that will become all too familiar:

% ./configure
% make
% make install

Network configuration, including assigning your machine’s name

and IP address, will follow the initial installation. Prior to placing your
system on a network, you need to consider the security of your sys-
tem. Programs such as tripwire can be set up after the initial install
and can be used to determine if files are being tampered with by an
intruder.

Starting Up / Shutting Down

Administrators need to consider how to start up and shut down the
system in a way that will not be disruptive to running processes and
users.

The administrator is responsible for configuration of the processes

that need to be started. UNIX runs through various run levels as the
system proceeds through the startup process.

The shutdown command is used to shut down or restart the system.

A conscientious administrator will warn his user base in advance of
system-wide activities like shutdown (using the wall command or
good old-fashioned email).

51
Users/Group Administration
Administrators are responsible for creating, updating, and deleting
users (using the commands useradd, usermod, userdel). Users are
categorized into groups, and administrators are also responsible for
maintaining these (groupadd, groupmod, groupdel).

System Monitoring
System administrators typically monitor servers to determine the
general availability of the system (the server is running and acces-
sible, file system space is available, etc). Web Application Develop-
ers need to provide additional information to system administrators
if they are expected to intelligently monitor application resources as
well.

Backup/Recovery
Perhaps the most important (and least appreciated) job of the
administrator is to perform backup procedures that guarantee that a
system can be recovered and data restored in the event of a system
failure. Archives can be created to back up essential files (tar, zip,
cpio) and the rsync utility can be used to mirror files on another
server. There are also a variety of proprietary tools for performing
such tasks.

Somebody’s Watching Me…

Monitoring is an activity that is often seen as the system adminis-
trator’s responsibility. However, web developers have a better under-
standing of their application’s features and requirements. A system
administrator might verify that a server is up or that a port is avail-
able—but without assistance from a programmer, they cannot (and
52
frequently do not) verify that all needed functionality is available.

System information about running processes can be viewed in the /

proc directory using cat. This directory contains special files which
provide information about the current state of the kernel. Some of the
relevant files that you might want to review:

/proc/cpuinfo
/proc/meminfo
/proc/version

On some systems you can use the procinfo command to get a

summary of this information.

To access a specific bit of information from one of these files, pipe the
results to grep and filter on the specific bit of information desired. For
example, to see Physical RAM:

% cat /proc/meminfo | grep MemTotal

Monitoring network resources can become necessary when develop-

ing multi-tiered applications. At minimum, you may need to ping
another server to verify its name/ip address and whether it is running
at all.

The fuser command can be used to identify processes using files

or sockets. The following command returns the pid of the web server
running on port 80:

% fuser -n tcp 80

The ps command can then be used to obtain more information about

the process or the nmap command can be used to verify that the port
53
is open:

% nmap --p 80 localhost

Other UNIX utilities (netstat, iptab, netcat) are available for net-
work administrative and security concerns.

The curl and wget utilities can be used in scripts used to monitor
web applications or copy web resources. They can be useful for creat-
ing web crawlers and automating web application testing. The follow-
ing is a simple example that could be used to populate a string after
a URL returned an HTTP status of 200. If an empty string was found,
an error would be reported.

#!/bin/bash
result=`curl -sIL localhost:8080/AFASWebConsole | grep ‘HTTP/1.1 200 OK’`
if [ -z “$result” ]; then
echo “ERROR: HTTP Response 200 expected.”
fi

Even though these utilities provide a good deal of functionality, more

complex tasks are better addressed through other software. Although
submission of forms, cookie management and other features are
available, browser automation is not directly possible (but is possible
using Watir, FireWatir, or Selenium). Parsing of HTML is better handled
using Hpricot or another parsing toolkit.

Command Description
ps Display process informa-
tion
pidof Returns the pid of a speci-
fied process

54
Command Description
top Display process informa-
tion interactively
vmstat Display memory statistics
report
free Display information about
system memory (e.g. RAM,
swap space)
du Display file space usage.
sudo du -hs /* will show
the directory sizes of the
root directories.
df Display file system space
usage
watch Periodically execute a
command
last List most recently logged
in users
lastlog List the last login time for
each user
ping Send an echo request to a
server
fuser Display a process using a
file or socket
nmap Display network hosts and
services
wget Non interactive download
of files from the web
curl Command line browser
typically used to copy
URLs

55
Monitoring File Space
Although monitoring the file space of a system generally is the
responsibility of a system administrator, you may need to monitor
specific file systems that are heavily used by your application. The
following example sends an email if the /home filesystem’s disk space
usage exceeds 90%. The script could be added to a user’s crontab
and scheduled to run on a periodic basis.

#!/usr/bin/bash
i=`df | grep /home | awk ‘{sub(“%”,””,$5);print $5 }’`
if [ “$i” -gt “90” ]
then
str=”`hostname` File System at $i”
echo $str | mail -s “`hostname` Alert” [email protected]
fi

This type of one-off script ceases to make sense when a larger

number of systems are in use. You can end up effectively spam-
ming yourself and receiving duplicate emails (for instance if several
web application servers report a database server being unavailable).
At that point, monitoring scripts might alert a centralized application
which sends emails out as appropriate.

Monitoring a Process
The ps command can be used to view running processes. To get all
information available (all processes and full listing) you can specify
a number of different options by including -ef. The resulting listing
can be limited by piping to grep. The process id (PID) is listed and is
helpful for other monitoring activities or stopping the process.

The top command can be used in an interactive mode to provide a

dynamic, real-time view of running processes that is somewhat akin

56
to window’s task manager. Various key strokes can be used to modify
the display while the program is running.

If you find yourself repeating the same command over and over (e.g.
running ls -l to check if a file has been created or changed), the
watch command can be used to periodically execute the command
in question. If a file is being changed by another program and you
would like to watch the file as it changes (a typical situation when
monitoring log files), the tail -f command can be used.

A job can be put into the foreground (fg), background (bg, Ctrl-Z,
&). A signal can be sent to a process—usually to stop it—using kill.
The kill -9 option sends a KILL signal that cannot be caught or
ignored. Because it does not allow for graceful termination, it should
be used sparingly.

Scripting
The Job That Ate My Brain
Normally we think about running programs by executing them at the
command line. You can start a process in the background by typing
an & after the command.

% command &

There are commands available to move a process between running

in the foreground or background. If you do start a process in the
foreground, you can stop the process in the background by pressing
Ctrl-Z and then out it in the background by typing bg. You can then
type jobs to see the job running in the background and fg to bring
the job back to the foreground.

57
Command Description
CTRL-Z Stop a running job
bg Put a job in the back-
ground
fg Bring a job into the fore-
ground
<command> & Run a job in the back-
ground
crontab View or modify the list of
scheduled jobs
at Execute a job at a speci-
fied future time
watch Execute a job periodically
and display the output to
the whole screen

Here are a few lines from a cron configuration file (the first line is
commented):

# Run script at reboot

@reboot sh /home/csaternos/bin/rebooted.sh
# Run every 30 minutes
# MIN HOUR MDAY MON DOW COMMAND
30 * * * * sh /home/csaternos/bin/file_system_alert.sh

review
• cron task (http://www.redhat.com/docs/manuals/linux/RHL-7.2-Manual/custom-
guide/cron-task.html)
• cron (http://www.unixgeeks.org/security/newbie/unix/cron-1.html)

Debugging
58
Shell scripting is an activity that you will naturally come to as you
develop sets of commands that can be organized into monolithic
tasks. Because the bash shell serves as an interactive interpreter, the
commands and syntax you are already using can be leveraged when
writing a script. This process is much like the process used by rubyists
to solve problems—experiment in Irb (the interactive ruby shell) and
commit final ideas to a script.

The environment available when running a scheduled script might dif-

fer from when it is invoked interactively. If this is the case, you might
want to invoke the script with the -l option which makes the script act
as if it were invoked as a login shell, or source the relevant files to set
up your environment.

• ~/.profile
• ~/.bash_profile
• ~/.bash_login
The simplest (and most common) technique for debugging scripts is
to use display output using echo and redirect output at various points
during a script’s output for subsequent analysis. The shell might be
invoked with a -v option (verbose) which prints shell input lines as
they are read. However, there are a options available that can mini-
mize these activities using the set built-in. The set -x option will
display each command as it is executed. This enables you to narrow
down what is happening at a given point in a script.

By default, if an error occurs, your script will continue processing. This

can be problematic if a command fails and its output was expected
by a subsequent command. The set -e option tells the script to exit
as soon as an error is encountered.

The set options can be specified within the script like so:

% set -ex
59
They also can be turned off (by changing the – to +) if you only want
to set behavior for a portion of the script.

They also can be set when invoking a script:

% sh -ex <script_to_debug>.sh

Scripts, like other programs, can suffer from performance issues. The
time command can be used to measure how much time a program
takes to complete.

% time <script_to_time>.sh

60
Appendix: Online Resources
chapter 8
linux
• The Linux Documentation Project (http://tldp.org/LDP/abs/html)
• UNIX Philosophy (http://catb.org/~esr/faqs/loginataka.html)
• FAQs (http://www.faqs.org/docs/artu/ch01s06.html)

utilities
• sed FAQ (http://www.linuxtopia.org/online_books/linux_tool_guides/the_sed_faq/
sedfaq4_005.html)
• top and tail (http://www.brunningonline.net/simon/blog/archives/002192.html)
• Network Settings (http://www.faqs.org/docs/linux_network/index.html)

bash scripting
• IBM Article Part 1 (http://www.ibm.com/developerworks/library/l-bash.html)
• IBM Article Part 2 (http://www.ibm.com/developerworks/library/l-bash2.html)
• IBM Article Part 3 (http://www.ibm.com/developerworks/library/l-bash3.html)
linux on oracle technology network
• Filtering (http://www.oracle.com/technology/pub/articles/saternos-filtering.html)
• Scripting (http://www.oracle.com/technology/pub/articles/saternos_scripting.html)
• Kickstart (http://www.oracle.com/technology/pub/articles/saternos_kickstart.html)

pipelines
• Wikipedia on Pipelines (http://en.wikipedia.org/wiki/Pipeline_%28Unix%29)

61
Appendix: Info Script
chapter 9
#!/bin/bash
echo “ “
echo “---------------------------------”
echo “- S E R V E R -”
echo “---------------------------------”
echo “ “
echo “Node name : `uname -n`”
echo “Machine name : `uname -m`”
echo “Operating System : `uname -o`”
echo “Kernel name : `uname -s`”
echo “Kernel version : `uname -v`”
echo “Kernel release : `uname -r`”
echo “Processor Type : `uname -p`”
echo “Hardware Platform: `uname -i`”
echo “ “
echo “---------------------------------”
echo “- M E M O R Y -”
echo “---------------------------------”
echo “ “
cat /proc/meminfo | grep MemTotal
echo “ “
echo “---------------------------------”
echo “- P R O C E S S O R S -”
echo “---------------------------------”
echo “ “
cat /proc/cpuinfo |grep “model name”| nl
echo “ “
echo “---------------------------------”
echo “- D I S K S P A C E -”
echo “---------------------------------”
echo “ “
df -h
echo “ “

62
63