Linux:

A comprehensive
introduction
Linux: A comprehensive introduction

How to Use this User Guide

This handbook accompanies the taught sessions for the course. Each section contains
a brief overview of a topic for your reference and then one or more exercises.
Exercises are arranged as follows:
 A title and brief overview of the tasks to be carried out;
 A numbered set of tasks, together with a brief description of each;
 A numbered set of detailed steps that will achieve each task.
Some exercises, particularly those within the same section, assume that you have
completed earlier exercises. Your teacher will direct you to the location of files that
are needed for the exercises. If you have any problems with the text or the exercises,
please ask the teacher or one of the demonstrators for help.
This book includes plenty of exercise activities – more than can usually be completed
during the hands-on sessions of the course. You should select some to try during the
course, while the teacher and demonstrator(s) are around to guide you. Later, you
may attend follow-up sessions at ITLP called Computer8, where you can continue work
on the exercises, with some support from IT teachers. Other exercises are for you to
try on your own, as a reminder or an extension of the work done during the course.
Text Conventions
A number of conventions are used to help you to be clear about what you need to do in
each step of a task.
• In general, the word press indicates you need to press a key on the
keyboard. Click, choose or select refer to using the mouse and clicking on
items on the screen. If you have more than one mouse button, click usually
refers to the left button unless stated otherwise.
• Names of keys on the keyboard, for example
the Enter (or Return) key are shown like this ENTER.
• Multiple key names linked by a + (for example, CTRL+Z) indicate that the first
key should be held down while the remaining keys are pressed; all keys can
then be released together.
• Words and commands typed in by the user are shown like this.
• Labels and titles on the screen are shown lik e t his .
• Drop-down menu options are indicated by the name of the options separated
by a vertical bar, for example File | Print . In this example you need to select
the option Print from the File menu or tab. To do this, click when the
mouse pointer is on the File menu or tab name; move the pointer to Print ;
when Print is highlighted, click the mouse button again.
• A button to be clicked will look lik e t his .
• The names of software packages are identified like this, and the names of
files to be used lik e t his .

11
Version 4.2 (TT 2019) 2 IT Learning Centre
 Linux: a comprehensive introduction

Contents
1 Introduction...................................................................... 6
1.1. Aims for Today............................................................................6
1.2. Course Outline............................................................................6
1.2.1. Session One...........................................................................6
1.2.2. Session Two...........................................................................6
1.2.3. Session Three.......................................................................7
1.2.4. Session Four..........................................................................7
2 Getting started..................................................................8
3 Command line exercises..................................................21
4 Editors, regular expressions, and shell scripts................35
5 Using remote computers.................................................48
6 Answers........................................................................... 57

IT Learning Centre 1 3 Version 4.2 (TT 2019)

1
Linux: A comprehensive introduction

Exercises
Exercise 1 A first look................................................................................................. 9
Exercise 2 Changing the keyboard layout.................................................................10
Exercise 3 The home area......................................................................................... 11
Exercise 4 Getting to know the desktop....................................................................11
Exercise 5 Finding your way around.........................................................................14
Exercise 6 Finding applications not on the Dock......................................................16
Exercise 7 Word Processing......................................................................................16
Exercise 8 Slide shows.............................................................................................. 17
Exercise 9 StackExchange........................................................................................17
Exercise 10 Starting a terminal window...................................................................17
Exercise 11 Where am I? What’s all this?.................................................................18
Exercise 12 File and directory manipulation............................................................20
Exercise 13 Viewing files.......................................................................................... 22
Exercise 14 Absolute and relative pathnames..........................................................23
Exercise 15 Help commands....................................................................................24
Exercise 16 File and directory names.......................................................................26
Exercise 17 Looking at files......................................................................................27
Exercise 18 Using wildcards to match filenames......................................................27
Exercise 19 Searching and sorting...........................................................................28
Exercise 20 Pipes and redirection.............................................................................29
Exercise 21 Finding the largest file..........................................................................31
Exercise 22 Merge information from different files..................................................31
Exercise 23 Unpacking a longer example.................................................................32
Exercise 24 Using gedit............................................................................................ 36
Exercise 25 Simple regular expressions...................................................................37
Exercise 26 Regular expressions with spaces...........................................................39
Exercise 27 Finding noodles.....................................................................................40
Exercise 28 Changing what you've found.................................................................40
Exercise 29 An example shell script..........................................................................42
Exercise 30 Developing your first shell script...........................................................42
Exercise 31 File manipulation scripts.......................................................................47
Exercise 32 Setting up access to IT Services Linux system......................................49
Exercise 33 Using ssh............................................................................................... 50
Exercise 34 Copying files between systems..............................................................51
Exercise 35 Copying directories between systems...................................................53
Exercise 36 Managing sessions on remote systems..................................................53

Version 4.2 (TT 2019) 4 IT Learning Centre

 Linux: a comprehensive introduction

1 Introduction
Today's course is divided into four parts each of which consists of a
presentation followed by exercises.
 Getting started: the desktop, office applications, the terminal.
 Using the command line.
 Editors, regular expressions, shell scripts.
 Working on remote computers.

1.1. Aims for Today

The course is designed to help you become a confident Linux user. The
topics covered are described in the Course Outline below.

1.2. Course Outline

1.2.1. Session One

This session will cover:
 A brief history of Linux.
 What is Open Source software?
 Who uses Linux.
 Office applications in Linux.
 What is a shell?
 Some simple commands.
The exercises will look at:
 Exploring and configuring the Linux desktop.
 Exploring LibreOffice applications.
 Simple use of the command line.

1.2.2. Session Two

As well as a powerful and useful desktop, Linux also has a command line
interface. This session starts to explore the command line. We will look at
 Viewing files, pathnames, getting help.
 Pattern matching
 Pipes or how to build your own commands
The exercises will cover:
 Using commands to manipulate files.
 Getting help.

IT Learning Centre 5 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

 Using patterns for searching for files.

 Building your own commands.

1.2.3. Session Three

This session builds on the previous session's introduction to the command
line. We will use the command line to do more complicated searches. We
will also look at shell scripting.
• Regular expressions.
• Text editors.
• Writing shell scripts.
The exercises cover:
• Using regular expressions patterns for searching.
• Building your own commands.
• Using an editor to develop shell scripts

1.2.4. Session Four

The final session looks at ways of using remote computers. It will also briefly
discuss managing your own computer.
• Remote access commnds.
• Using package managers to find and install software.
The exercises will cover
• Using ssh to access a remote computer
• Using scp to move files and directories between computers
• Using screen to leave a program running on a remote computer after
disconnection

Version 4.2 (TT 2019) 6 IT Learning Centre

 Linux: a comprehensive introduction

2 Getting started
Here are some things to try to help you get to know Ubuntu Linux.

Exercise 1 A first look

This is the standard Live Ubuntu start-up screen.

Let's look at the screen more carefully. The bar along the top has three components:

Show current applications and search for others

Time – can be configured to include date

Various settings:
• Language and keyboard settings
• Volume control
• Battery
• Systems settings and logout

On the left side of the Desktop, below the Activities launcher is the Dock.

IT Learning Centre 7 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

Under the Dock you should see a vertical list of popular applications. If you hold your
mouse over any of these icons then their name pops up.
From the top these are:

Install Ubuntu Install Ubuntu on your PC. Don't use this now!

Firefox Web Browser Browse the web

Thunderbird Mail Read email

Files Find files and folders and explore the system

Rhythmbox Music player

LibreOffice Writer Create documents, similar to Microsoft Word

Ubuntu Software Find, add, update and remove applications

Help Get help

Link to Amazon

The Install icon can be used to install Ubuntu on your PC, but please don’t do this!
At the bottom of the Dock is

Show Applications

Exercise 2 Changing the keyboard layout

The keyboard is currently set up with the US-style layout. To set it to a UK layout
i. Right click on en1 on the top right bar

Version 4.2 (TT 2019) 8 IT Learning Centre

 Linux: a comprehensive introduction

ii. Select English (UK) en3

You should now be using a UK Keyboard layout.

Exercise 3 The home area

When you log onto a Linux system you are located in your home area. This is where
any files that are created during this session are saved. See if you can find a quick way
to open a window which shows the contents of your home area.
The answers are at the end of the exercises.

Exercise 4 Getting to know the desktop

I. Finding applications
From the Dock find
• The Firefox Web Browser
• LibreOffice Writer

IT Learning Centre 9 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

• Help
From the top panel find
• System Settings [Hint: Look for the crossed wrench/screwdriver icon
under the arrow on the top right]

II. Window operations

We're going to experiment with actions on windows. Start Firefox. Make sure
you can do the following to this window
i. Maximise – make full screen size.
ii. Minimise – close the window and send to the Dock.
iii.Restore – bring back a minimised window from the Dock.
iv.Close – exit from a window permanently.
v. Resize – make the window smaller or larger.
vi.Display options for a window. [Hint: right click on the dark grey bar at the
top of a window.]
Did you notice that when you have started an application there are small orange
dots to the left and right of the icon on the Dock? What happens if you click on a
different application? Did you notice that the appearance on the Dock of the
active application changes? What happens if you open a new copy of the
application so that there are two (or more) instances running?

III. Configuring your desktop

System Settings allows you to configure your desktop environment and many
other features.
i. Open the System Settings menu and click on Background.

Version 4.2 (TT 2019) 10 IT Learning Centre

 Linux: a comprehensive introduction

ii. Change your wallpaper. Ubuntu Live has a limited range of wallpapers; a full
install will have more.
iii. Linux supports the use of multiple workspaces or desktops. This allows you to
organise work more efficiently and reduce clutter. Press the Super (or
Windows) key on the bottom left of the keyboard:

to display additional desktops. You should see something like this:

You can then use the mouse to move between workspaces, and move
applications between these workspaces.

IT Learning Centre 11 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

In this screenshort there are applications open in two workspaces.

iv. More difficult: Change some keyboard shortcuts. Keyboard shortcuts use the
keyboard rather than the mouse to perform actions. Go to Keyboard in the
Devices page of System Settings) and use this to set up a quick way to switch
between workspaces. Scroll through this list of shortcuts, looking for Switch to
workspace 1 and replace Super+Home by, for example, CTRL+1 [Hold the
CTRL key and 1 down at the same time]. Switch to workspace 2 to CTRL+2
and so on. Do this for 4 workspaces. You should now be able to switch between
workspaces using these keystrokes. Some keyboard shortcuts are pre-
configured.

Exercise 5 Finding your way around

Go to the Dock and open Files.

Version 4.2 (TT 2019) 12 IT Learning Centre

 Linux: a comprehensive introduction

You can use Files to browse files and directories (the Linux equivalent of folders).
You should be in an area called Home. When you start the file browser, it opens in
your home directory. This is the place where your files will be stored by default – that
is if you don't specify another location.
Click on the small grid on the top bar

The view in Files should change

You can change the properties you see for each file and directory by right clicking
Name above the list of directories. A list of properties appears:

IT Learning Centre 13 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

Change this list so that you can see the Type, Owner, Permissions and Modified –
Time.
Finally, see if you can find a directory that you can't see inside. We will talk about how
privileged accounts are used in the final session.

Exercise 6 Finding applications not on the Dock

There are only a limited number of applications on the Dock. What if you want to use
something that isn't there? Clicking on the Activities icon displays a window like this:

To find an application you can either browse for it by category or search in the search
box. In later sessions we will be needing a terminal window. Enter terminal in the
search box and see what happens. You may not need to type in the whole word to get
what you want.
Once you have found what you want, a single click on the icon will open the
application. As well as opening the application, an icon should also appear on the
Dock. If you right click on this icon, you can keep the application there by selecting
Add to Favorites, even after you have closed it. Test this with the terminal window.
The following two exercises look at the Libreoffice equivalents of Microsoft Word and
Microsoft Powerpoint. As time is short, choose the one you are most likely to find
useful.

Exercise 7 Word Processing

Start LibreOffice Writer from the Dock.
Write a short document (perhaps a covering letter for a job application for example).
Don't spend too long on this but see if you can find out (from your experience with MS
Word or similar) how to insert things like tables, do indentation and so on. If there's
something you'd like to see but you're not sure where it is, ask for help.
Now save what you've done. By default the file is saved in ODF (Open Document
Format). Once you done this, see if you can find out how to save a document in MS
Word format instead. You can also save the file in PDF format. Again see if you can
find out how to do this (it's not in quite the same place).
Now let's close this document by clicking on File and Close.
Use your Firefox browser to visit http://www.stats.ox.ac.uk/pub/susan/linux/
You should now see an MS Word file, Usingmsc.doc, available for downloading. Save
this file (right click on the link) and make sure it has downloaded by checking your
Home Folder Window. Go back to the word processor and open this file. It should
work perfectly even though it was created with MS Office.

Version 4.2 (TT 2019) 14 IT Learning Centre

 Linux: a comprehensive introduction

Although most MS Office files work with LibreOffice, a few don't convert very well so
be warned that it's not 100%compatible. (Maybe 98% is a fair number).
Now close Writer and go to the StackExchange exercise.

Exercise 8 Slide shows

Use LibreOffice Impress to make your slide show; it is an application similar to MS
Powerpoint. You will need to use Activities to search for this.
You don't have time to make a large presentation today so we suggest as an exercise
that you make one or more of:
• A three page presentation about your course/research
• A poster for an event/party
• A flier for Ubuntu Linux
Use the Firefox browser in Desktop 1 to search for and to download images and insert
them into your presentations. Ask your demonstrators if you can't find the features
you want to use.
Save your work as both in Open Document format and MS Powerpoint format.
Now close the application.

Exercise 9 StackExchange
We will now look at another source of information and help that is available.
StackExchange provides a gateway to communities of expertise. A particular benefit of
the sites is the lack of additional chat – in general there is a strong focus on direct
answers to questions. Browse to http://stackexchange.com/about to find out more.
Browse to http://www.stackexchange.com and click on 'Explore our sites!' to see the
huge range of topics covered. Let's investigate the 'Ask Ubuntu' site. On my browser
the link to this site appears in a large orange box to the right of the large
'Mathematics' box which is near the top on the left hand side. If you can't find the link
then use the browser search feature (usually CTRL+F) and enter Ubuntu.
When you click on the 'Ask Ubuntu' box you should see a 'Visit Site' button. Click on
this.
Now click on 'Take the 2-minute tour' to see a brief guide.
If, at the end, you have time have a look at other useful sites such as 'Unix &
Linux'.

Exercise 10 Starting a terminal window

Now that you are more familiar with the Ubuntu desktop, let's start looking at the
command line interface.
The first thing we need is a shell prompt/command line. When using a graphical
desktop there is usually a terminal application which gives you access to the command
line.

IT Learning Centre 15 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

Click on 'Activities' and enter Terminal in the search box. When you have found the
terminal open it. If you want to keep it on the Dock, right click on the icon on the left
hand side of the screen and select 'Add to Favourites'.
You should see a Window open that looks like this:

Note that it's perfectly possible to have a command line with no graphical desktop at
all. This is often the case with server systems which are not used interactively and
need all the processor power and memory they can get for computation.
We will be finding out more using remote server systems in the final section.

Exercise 11 Where am I? What’s all this?

Let's start to look at navigation of the Linux file system. The following commands are
introduced:

Command Purpose

pwd print working directory. In other words,

“where am I?”

ls [options] directory List files. If used on its own, it lists

everything in the “current working
directory“ (where you are currently
“located”).

file filename Tells you what sort of file the file called
filename (for example) is.

cd Change directory. In other words, “please

change my current location”.

man command Command manual pages.

Version 4.2 (TT 2019) 16 IT Learning Centre

 Linux: a comprehensive introduction

Note that all commands are typed in lower case. There are very few Linux commands
which have any uppercase (CAPITAL) letters. We will look at case-sensitivity and file
names in the next session.
Right away we can see how quiet Linux commands are by default. Try typing in
cd
at the ubuntu@ubuntu:~$ prompt and you will get no output at all. This does not mean
that anything has gone wrong. For many commands, no output means successful
completion.
A digression on prompts. You can customise your prompt to look however you like. We
won't do that now, but you will notice that it changes as you move around the file
system.
Not all commands are silent. Try
pwd
You should get a response like: /home/ubuntu. Now try
ls
You should now see a listing of all the files in the directory /home/ubuntu. Let's try
finding out what sort of files each is. Take the file “Desktop”.
file Desktop
The shell tells you that this isn't a regular file, it's a directory. In other words it's a
special file which acts as a holder for yet more files (like a folder in Windows). If you
now try
file Desktop/*
you should see a description of the two files in that directory.

TIMESAVING INFORMATION
If you haven't already tried this then you should now. A lot of typing can be avoided by
several useful shortcuts. The <tab> key can be used to complete commands file
names and the arrow keys to recall previous commands and perhaps change them.

IT Learning Centre 17 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

Exercise 12 File and directory manipulation

Now you are going to create a directory and put some files there. The commands you
need are

Command Purpose

cd change directory. In other words,

“please change my current location”.

mkdir directoryname create a directory called

directoryname

touch file1 file2 create one or more empty file(s)

called file1, file2

cp file1 file2 copy file1 to file2. Can also be used

to copy whole directories.

ls List files. If used on its own, it lists

everything in the “current working
directory“ (where you are currently
“located”).

rm file1 remove (or delete) a file called file1.

Can also be used to remove whole
directories.

cd
mkdir directory1
cd directory1
touch file1 file2 file3 file4
Remember that words in italic should be replaced by names that you have chosen.
Experiment to see what happens if you are not in your home directory. What happens
if you try to create a directory in /usr/bin? Is there anywhere outside your home
directory where you are allowed to create directories? [Hint: look at the top level
directory / - you should be able to create a files and directories in one of those. The
name of the directory might also be a clue. Check in the Answers section below.]
Use the cp command to copy one file to another and then use ls to check that you
have done what you want. Then delete a file using
rm file1
Now we are going to copy one directory to another. The commands you need are
cd
cp -r directory1 directory2

Version 4.2 (TT 2019) 18 IT Learning Centre

 Linux: a comprehensive introduction

Use ls to make sure you have done what you want. The new directory should contain
exactly the same files as the old one. Note use of the -r option. This makes cp copy
the contents of a directory – this is known as a recursive copy.
Finally remove the new directory with
rm -rf directory2

Note that this is a dangerous command and should be used with care!
Use ls to check that this has worked. You should now be familiar with these simple
file manipulation commands. Remember that in Linux the rm command really does
delete files. There is no Recycle Bin to retrieve files that were deleted by mistake.

THIS COMPLETES THE EXERCISES FOR SESSION ONE

IT Learning Centre 19 Version 4.2 (TT 2019)

 Linux: A comprehensive introduction

3 Command line exercises

Exercise 13 Viewing files
We're going to download some files and directories which will be used during these
exercises. Although it is possible to use a browser to download this file you can also do
this from the command line. Use
cd
wget https://www.stats.ox.ac.uk/pub/susan/linux/LinuxFiles.tgz1
to download the files and then
tar -xvzf LinuxFiles.tgz
to unpack them.

Command Purpose

cat file show the whole contents of a file

called file

more file display the contents of file a

screenful at a time

less file display the contents of file a

screenful at a time but with more
options. For example, after starting
less you can enter G and go straight
to the end of a file and then move
backwards.

Use the following commands to look at the contents of the file google.txt.
cd Files
cat google.txt
This is not very useful if the file is more than a screenful.
more google.txt
Note that <space> takes you to the next page and q will quit before the end of the
file. Now try
less google.txt
See if you can go straight to the end of the file. Then use q to exit.

1
This command is not available on MacOS. Use curl instead. For example
curl -O http://www.stats.ox.ac.uk/pub/susan/linux/LinuxFiles.tgz

Version 4.2 (TT 2019) 20 IT Learning Centre

 Linux: a comprehensive introduction

Exercise 14 Absolute and relative pathnames

We're now going to make use of two things, the ls command and the knowledge that
the file called /usr/share/info is a directory, to illustrate the concepts of absolute
and relative pathnames.
cd
cd ../../usr/share
ls info
and you will get a listing of the contents of the directory.
ls /usr/share/info
and you should get the same list of files.
The absolute (i.e. complete) location of the info directory is /usr/share/info. We
have just asked to see what is kept inside it in two different ways. The first is a
relative pathname while the second is the full or absolute name.
Imagine the info directory is a particular house, say 42, High Street, Abingdon and I
ask you to deliver a letter there. I could tell you to deliver the letter to “ 42 High
Street, Abingdon”: the full/absolute address. No matter where you are in the UK,
that's enough information. However, if you were already in Abingdon I could tell you
to deliver the letter to the relative address of “ 42, High Street” or even better, if you
were standing on the high street just “number 42” would be enough.
The “ls info” command above worked because you were already in the /usr/share/
directory. It wouldn't work from somewhere else. The command “ls
/usr/share/info” command will work from anywhere (although it's more long
winded). Let's prove it by changing our current location using the cd command to go
back to the home directory
cd
cd Desktop
pwd
you should get “/home/ubuntu/Desktop” i.e. you have moved into the Desktop
directory.
ls /home/ubuntu/Desktop
should give you the list of files in that directory. In fact you could use “ls” on its own
without the name of the directory because you have already moved there with cd.
Let's see what happens when we deliberately do something wrong:
ls Desktop
should give you an error saying there is “No such file or directory” which is correct.
The command fails because Desktop on its own is a relative name and you've started
from the wrong place.
Let's expand the idea of relative and absolute path names using the cd command.
Make sure you are still in the Desktop directory before you start (check with pwd).
pwd
cd ..

IT Learning Centre 21 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

pwd
cd ..
pwd
and so on until you can't go any further (you won't see an error, you just stop going
anywhere). “..” is a special location which means “up one level”. All directories
contain a “..” so you can go up a level. The exception is called “/” or “the root” or
just “slash”. You can't go higher than / so “..” doesn't take you anywhere. Note that
there is another special directory called “.” which means “current location”. All
directories contain a “.” directory and we'll see why it is needed later.
During the above task you went up the directories one level at a time. Now reverse the
process and go back to the Desktop directory one level at a time. You should be in “/”.
Note that you don't have to do the pwds but it may help you visualize what is going on.
You can also use ls to have a look around each level if you have time.
cd home
pwd
cd ubuntu
pwd
cd Desktop
pwd
Try to answer/do the following:
Were you just using absolute or relative paths?
1. Now try to get back to the root (or “/”) directory with one command only using an
absolute path.
2. Now get back to the /home/ubuntu/Desktop directory using one command only.
3. What are the contents of the “/” directory? From your home directory use one
command only to find out.

Exercise 15 Help commands

Command Purpose

man command Read the manual page for a

command. So man ls gives you
details about ls and man less would
give you details about less.

apropos word Search manual pages for names and

descriptions. So apropos copy
would list all the commands that
have the word copy in the
description.

Version 4.2 (TT 2019) 22 IT Learning Centre

 Linux: a comprehensive introduction

Command Purpose

which command Displays the location of the

command you are using.

whatis command Gives a brief description of a

command.

If you know what command you need, you can use the man command to find out the
details of that comand. Try it with a few of the commands you have used already. Not
all commands have as many options as ls!
man ls
to find out details of the ls command.
1. What option is used to display modification time?
2. What option is used to display the size of a file?
3. How can you reverse the order of the sort so that the largest/most recently
changed file is at the bottom of the list?
4. Check that they do what you expect.
Sometimes you might not be sure exactly what the command is. In that case you can
use the apropos command which finds all command descriptions which match a given
word. So to find out what commands there are to manipulate files are available use
apropos file
Note that the output from this command is very long. We will look at a neat way round
this in the next session. Here is a real life example. I needed to know which command
in this Ubuntu distribution was the equivalent of Microsoft Draw so I ran
apropos draw
See if you can see the name of the command I used. The answer is at the end of the
exercises.
Sometimes you need to know where Linux stores command. Use which to display the
location of the file. Try it with less, more, cp, apropos:
which apropos
which cp
which more
which less
Did you notice that more and less are stored in different places? The directory /bin
conventionally is used to store a few necessary commands that can be used if all else
fails. Finally you may have seen a command and want to know briefly what it does.
Use the whatis command to find out. Try this on some commands.

IT Learning Centre 23 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

Exercise 16 File and directory names

These are the commands that will be used:

Command Purpose

cd change directory. In other words,

“please change my current location”.

cat file show the whole contents of a file

called file
We’re going to look at some of the problems you can encounter with files and
directories. You should have a directory called Files. Now do this
cd
cd Files
cd TestDir
First look in the Cases directory.
cd Cases
Make sure that you can read all three files there. Now look in the OpenThis directory.
cd ..
cd OpenThis
1. Now see if you can read the files star, astar and *star. What happens when you
try to read the file called *star? The * character has caused confusion because it is
a wildcard – it matches any character. We will be looking at wildcards and pattern
matching in a later exercise. See if you can read this file without reading the other
ones. Using google is not cheating!
2. Now change into the directory Open This. Double quotes or backslashes help.
3. Now try to read the file called -ReadMe. Using quotations round the filename or the
escape character (\) won’t work this time. In each case the command (one of cat,
more or less) is interpreting the leading – (hyphen) as an option. Again, don’t be
afraid to google to see if you can find the answer to this. Try deleting the file called
-DeleteMe. To help you with this, man less says:
-- A command line argument of "--" marks the end of option arguments
Any arguments following this are interpreted as file‐names. This can
be useful when viewing a file whose name begins with a "-" or "+".
See the answers for some suggested solutions.

Version 4.2 (TT 2019) 24 IT Learning Centre

 Linux: a comprehensive introduction

Exercise 17 Looking at files

These are the commands that will be used:

Command Description

cat file Show the entire contents of a file called file.

head file Display the first 10 lines of a file.

less file Display the contents of file a screenful at a time but with
more options. For example, after starting less you can
enter G and go straight to the end of a file and move
backwards.

man command Read the manual pages to read all about a command. So
man head
would describe how to use the man command.

more file Display the contents of file a screenful at a time.

tail file Display the last 10 lines of a file.

wc file Counts the number of characters, words and lines in a file.

Using
cd Advanced
cat longfile.txt
you see all of longfile.txt; using
less longfile.txt
gives you a screenful at a time. Use the spacebar to move on a screenful, G to go to the
end of the file and q to exit. Now use
head longfile.txt
and
tail longfile.txt
to look at the first and last 10 lines of the file.
Now use the wc command to find out the number of lines, words and characters there
are in
longfile.txt.

Exercise 18 Using wildcards to match filenames

File globbing or wildcard expansion allows you to use special characters to match
more than one file or directory name.

IT Learning Centre 25 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

Character What it matches

* The * (asterisk or star) matches any number of characters

or none.

? The ? (question mark) matches exactly one occurrence of

any character.

[ ] Matches any characters in a given range.

Change back to your home directory (if you are not already there) and then change to
the directory called WildCards.
cd
cd WildCards
Now experiment with wild card characters. What do the following match?
ls foo?
ls foo2*
ls foo[1-2]
What command would you need to match just foo20 and foo2bar? [Hint: you might
need to use more than one wild card character.] The answers are at the end. Now use
wc -l *
to find the length of each file. Note that the -l is a hyphen followed by the lowercase
letter l. You should see output like this:
1 foo
2 foo1
2 foo10
1 foo2
1 foo20
3 foo2bar
10 total
Now see if you can create a match so that wc -l just shows the files with a 1 (the
number one) in their name. Again a possible answer is at the end.

Exercise 19 Searching and sorting

In this exercise we will use commands to search for patterns within files and sort the
contents of files.

Version 4.2 (TT 2019) 26 IT Learning Centre

 Linux: a comprehensive introduction

Command Description

grep pattern file Search for the characters in pattern within

file.

sort file Sorts the contents of one or more files.

Now change into the Searching directory.

cd
cd Searching
I suggest you use the cat command to check the contents of the two files, fruit and
veg, so that you can see what they contain. Now use
grep melon fruit
Can you see what has happened? Only the matching lines in the file are shown. Try
grep green fruit veg
What do you think has happened here? Did you notice that the file names were
included in the output. This is because more than one file (both fruit and veg) has
been searched and so the output informs us where the matches were found.
Using
man grep
see if you can find the option that causes grep to ignore case so that both Melon and
melon would be found.
Finally use the sort command to order the files. Note that you can sort more than one
file which merges the output of all the files.
sort fruit veg
What is the option that reverses the order – so that the fruit and vegetables are sorted
in reverse alphabetical order?

Exercise 20 Pipes and redirection

In this exercise we are going to explore two very powerful command-line features
which increase the flexibility and range enormously.
Command Description

du -sk Displays sizes in Kilobytes of all files in a directory.

grep Search for the characters in pattern within file.

sort Sorts the contents of one or more files.

tail View the last few lines of a file

wc Counts the number of characters, words and lines

in a file.

IT Learning Centre 27 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

and we will be using the following characters. If you haven't already remapped your
keyboard you should do that now. See Exercise 2.

Character Purpose

> Sends the output from a command to the named

file. If the file already exists the previous contents
will be lost. If the file doesn't exist it will be
created.

>> Appends the output from a command to the named

file. If the file doesn't exist it will be created.

< Reads input from the named file. NB This option is

rarely used.

| Uses the output from one command as the input to

the next.

Almost all Unix/Linux commands use standard input for receiving instructions and
standard output for displaying the results. So we could run
grep green fruit veg > output
which would store the output of the grep command in a file called output. Check this
with
cat output
If you run the command again, this time using
grep green fruit veg >> output
you should see that there are now two copies of the output. It is also possible to use <
redirect the input from the keyboard to a file. For example
cat <output
also works.
In the previous exercise we used grep to look for occurrences of the word green in
two files. Obviously in this short example we can count how many times green
appears, but when there are many matches it would be useful to use wc to find out. We
can redirect the output from the command into a file.
Note that in all cases the -l option to the wc command is a minus sign followed by the
lower case letter l.
grep green fruit veg >output
and then run wc on the file
wc -l output
but it would be much more efficient to join the two commands together with a pipe.
Use

Version 4.2 (TT 2019) 28 IT Learning Centre

 Linux: a comprehensive introduction

grep green fruit veg | wc -l

Here the two commands grep green fruit veg and wc -l are joined together by a
special symbol called a pipe.

Exercise 21 Finding the largest file

Now we're going build a longer command which will find the 5 largest files in a
directory. When building pipes of commands it often helps if you make sure each link
in the pipe works before adding the next.
First display the size in Kilobytes of all the files in the /usr/bin directory.
du -sk /usr/bin/*
Now sort this output by size, so that the largest are first.
du -sk /usr/bin/* | sort -nr
Now display the final 5 lines which will be the 5 largest files.
du -sk /usr/bin/* | sort -nr | head -5
Now how you would you find the 5 largest files in the /usr/bin directory beginning
with the letter 's'?

Exercise 22 Merge information from different files

Command Description

awk A pattern scanning and text matching program.

paste Merge lines of a file

This is a rather contrived example but demonstrates some very useful Unix/Linux
commands. Make sure you are still in the Searching directory.
cd
cd Searching
Now we are going to use the awk command to print out only the first column of the file
called creatures. Have a look at the file before you run this command.
awk '{print $1}' creatures
What do you think would appear if you replaced $1 by $2? Make sure you copy the
command exactly. In particular you need the single quote character and curly
brackets. Now change this to
awk '{print "A", $1, "eats"}' creatures
As well as a list of animals you should also see that the words “A” and “eats” appears
either side of each animal. Now we are going to use the paste command to include the
output from the fruit file.
awk '{print "A", $1, "eats"}' creatures | paste -d" " - fruit

IT Learning Centre 29 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

Again, make sure that you have entered the commands exactly as they appear here. To
explain what the paste command is doing: the -d” “ option makes a space rather
than a tab appear before the name of each fruit, the – before fruit means that the
output from awk... is included.
Now see if you can make the fish eat the vegetables!

See the Answers section below for an answer.

Exercise 23 Unpacking a longer example

In this exercise you will look at a longer example and repeatedly remove commands to
unpack a complex command. Enter these commands.
cd
cd Advanced
cat my_wordcount

We will be using these commands

Command Description

sed sed allows you to change the contents of a file; frequently

used in pipes.

sort Sorts the contents of one or more files.

tail View the last few lines of a file

tr Translate either groups of characters or single characters.

uniq Report or remove duplicate lines.

When you run the commands

chmod +x my_wordcount
./my_wordcount longfile.txt
you should see output like this:
6 microsoft
6 more
6 mr
6 said
6 that
7 has

Version 4.2 (TT 2019) 30 IT Learning Centre

 Linux: a comprehensive introduction

7 yahoo
8 engine
9 and
9 on
10 pages
10 web
11 in
12 of
13 its
14 a
15 search
16 google
20 to
29 the

This is a list of the 20 most common words used in the file longfile.txt. Now, from
the command line, we’re going to run the command repeatedly, each time removing
the final element, to unpack how it works.
Using the mouse, copy the contents of my_wordcount and paste into the command
line. You should have a very long command like this:
sed -e 's/\.//g' -e 's/,//g' -e 's/\!//g' -e "s/\'//g" -e "s/\"//g" -e
's/-//g' longfile.txt | tr ' ' '\012' | tr [:upper:] [:lower:] | sed -e
'/^$/d' | sort | uniq -c | sort -n | tail -20
This looks very complicated!

Now lets remove the final pipe which is

| tail -20
so you have
sed -e 's/\.//g' -e 's/,//g' -e 's/\!//g' -e "s/\'//g" -e "s/\"//g" -e
's/-//g' longfile.txt | tr ' ' '\012' | tr [:upper:] [:lower:] | sed -e
'/^$/d' | sort | uniq -c | sort -n
Remember that you can use the up arrow and backspace keys to do this. There is no
need to type anything in. This time the output shows all the words in the file, not just
the 20 most frequent. Now let's remove
| sort -n
and see what happens
sed -e 's/\.//g' -e 's/,//g' -e 's/\!//g' -e "s/\'//g" -e "s/\"//g" -e
's/-//g' longfile.txt | tr ' ' '\012' | tr [:upper:] [:lower:] | sed -e
'/^$/d' | sort | uniq -c

IT Learning Centre 31 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

Now the list is sorted alphabetically rather than numerically. Again, remove the final
pipe,
| uniq -c
and see what happens.
sed -e 's/\.//g' -e 's/,//g' -e 's/\!//g' -e "s/\'//g" -e "s/\"//g" -e
's/-//g' longfile.txt | tr ' ' '\012' | tr [:upper:] [:lower:] | sed -e
'/^$/d' | sort

Now you should see an alphabetical list of all the words, but with multiple occurrences
of many words.
Remember you can always add
| less
on to each version of the command so that you can look at the output a screenful at a
time.
The uniq command (which is very useful) removes duplicates, and with the -c option,
adds a count of the number of times it has found a match. Now remove the final pipe,
| sort
again so that you have
sed -e 's/\.//g' -e 's/,//g' -e 's/\!//g' -e "s/\'//g" -e "s/\"//g" -e
's/-//g' longfile.txt | tr ' ' '\012' | tr [:upper:] [:lower:] | sed -e
'/^$/d'

This time you have all the words in the file with one word on each line but in the order
they appeared in the original file, not sorted alphabetically. You can read the contents
of longfile.txt if you want to check this. Now again remove the last pipe again
| sed -e '/^$/d'
and see what the output looks like.
sed -e 's/\.//g' -e 's/,//g' -e 's/\!//g' -e "s/\'//g" -e "s/\"//g" -e
's/-//g' longfile.txt | tr ' ' '\012' | tr [:upper:] [:lower:]

The file no longer contains any blank lines. If you haven't already add | less on to
this version and the previous version to compare. Again, remove the final pipe,
| tr [:upper:] [:lower:]
and see what happens.
Did you notice that some uppercase letters appeared? The tr command allows you to
translate either classes of characters (such as all uppercase characters) or individual
characters to something else. In this case we are translating all uppercase characters
to their lowercase equivalent. Now remove the final pipe again.
sed -e 's/\.//g' -e 's/,//g' -e 's/\!//g' -e "s/\'//g" -e "s/\"//g" -e
's/-//g' longfile.txt | tr ' ' '\012

Version 4.2 (TT 2019) 32 IT Learning Centre

 Linux: a comprehensive introduction

Now you should see the file with many words on each line. In this case we have used
tr to translate all space characters to the newline character (which is represented by
\012).. The first command – which is very complicated - removes some punctuation
from longfile.txt.
sed -e 's/\.//g' -e 's/,//g' -e 's/\!//g' -e "s/\'//g" -e "s/\"//g" -e
's/-//g' longfile.txt

This is a complicated command! We are using the sed editor to match and remove
several separate things. Each match begins with -e and involves replacing various
punctuation characters with nothing at all. In some cases the punctuation character is
preceded by “\” because it has special meaning in the sed editor.

THIS COMPLETES THE EXERCISES FOR SESSION TWO

IT Learning Centre 33 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

4 Editors, regular expressions, and shell scripts

Exercise 24 Using gedit
First open a terminal window so that you can use the command line. Using this
window type in
gedit newfile &
and a window like this should appear. This command starts the gedit editor, creates a
file called newfile and returns you to the shell prompt.

The new gedit window you have opened consists of

• a menu bar at the top with Open, New Document and Save buttons. Note that an
expanded menu appears on the bar at the top of the screen.
• a main window for entering text
If other files are open then you can have a series of tabs in the main window.
You can now enter text in the main window. You can move around the file with the
arrow keys or the Page Up and Page Down keys.

Version 4.2 (TT 2019) 34 IT Learning Centre

 Linux: a comprehensive introduction

Try them! Now see if you can enter some text into the file, save the text and quit gedit.
Now start the editor up again with the same file, change the text, save it and quit
again.

A small digression. When you started gedit you entered gedit newfile &. The &
character has a special function. It allows you to start a command and continue using
the terminal window to enter further commands. Technically it is called 'running a
program in the background'. See what happens if you don´t include the & - just type in
gedit newfile. You will need to exit gedit or enter CTRL+C [ie press the CTRL key and
C at the same time] to get back control of the command line.
A brief summary of some frequently used command line control sequences are
CTRL+C interrupt a running program
CTRL+D send an end of file, ending text input for most Linux/Unix programs
CTRL+Q unfreezes the screen if CTRL-S has been used to freeze it.
CTRL+U deletes the last line typed
CTRL+Z suspends a running program. Use bg to continue running the program in
the background or fg to continue running the program as it was.

Exercise 25 Simple regular expressions

Or finding needles in haystacks! We are going to use the following commands and
shorthand characters..

Command Description

gedit A simple text editor

grep Print lines matching a pattern.

Options used:
-n display line numbers
-E use extended regular expressions

Regular expression Description

shorthand
character

^ Match the beginning of the line

$ Match the end of the line

Have a look at the file HAYSTACK. It's going to be easier to look at it in gedit so enter
cd

IT Learning Centre 35 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

gedit HAYSTACK &

from a terminal window. If the line numbers are not displayed it is helpful to show
them. To do this, on the top panel, expand the TextEditor menu:

and select Preferences. On the View tab select ‘Display line numbers’ which should be
the first item.
We'll spend more time looking at gedit in the next session, but for now we're going to
use it to show the contents of a file.
Have a look at the file. There are many lines of text with mostly the word 'hay' none or
more times. There are some needles hidden in there, as well as some nedles, neeedles
and noodles.
A simple search
grep -n -E needles HAYSTACK
will show all the occurrences of needles in the file.
grep -n -E ^needles HAYSTACK
will show the lines that contain needles right at the beginning of the line. The search
grep -n -E needles$ HAYSTACK
will show only the lines that have needles at the end of the line.
How would you search for a line which contains only one occurrence of needles and
nothing else?
See the Answers section to find out.

Version 4.2 (TT 2019) 36 IT Learning Centre

 Linux: a comprehensive introduction

Exercise 26 Regular expressions with spaces

Command Description

grep Print lines matching a pattern

Regular expression Description

shorthand character

\s match white space

\b match a word boundary

\B match a non-word boundary

+ matches one or more of the previous character

Sometimes words at the beginning of a line are preceded by white space (either a tab
or a space). Regular expressions have many character shorthands which allow us to
make general searches. In this example we use '\s' to match a single white space
character.
grep -n -E '^\sneedles' HAYSTACK
to find all the lines that begin with a single white space and then needles.
What if there are multiple spaces? We can use the '+' character which will match one
or more instances:
grep -n -E '^\s+needles' HAYSTACK
How many lines end with one or more white spaces?
grep -n -E '\s+$' HAYSTACK | wc -l
Note that this time, we haven't restricted the search just to lines ending in needles.
In some places hay and needles have been run together. To find only the places where
needles is separate from hay use
grep -n -E '\bneedles\b' HAYSTACK
This is a particularly useful search because it means we don't have to worry about
matches for beginning and ends of lines. Incidentally we can reverse this by using the
grep -n -E '\Bneedles' HAYSTACK
and
grep -n -E 'needles\B' HAYSTACK
which will find instances where needles are joined to hay either at the beginning or
end of the word.
What happens if you use
grep -n -E '\Bneedles\B' HAYSTACK
See the Answers section.

IT Learning Centre 37 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

Exercise 27 Finding noodles

Command Description

grep Print lines matching a pattern

Regular expression Description

shorthand
character

| give alternative matches

. matches one or more characters

{} specifies a number of matches

Looking at the contents of HAYSTACK, we can see that there are words similar to
needles - noodles, nedles and neeedles. How can we find them all?
We could use the alternation character: '|' to find needles or noodles.
grep -n -E 'needles|noodles' HAYSTACK
This is a little restricted as we may need to specifiy many alternatives. Try this:
grep -n -E 'n.{0,2}dles' HAYSTACK
This will match all strings that
• start with 'n'
• have 0, 1 or 2 characters before 'dles'

Exercise 28 Changing what you've found.

This is slightly more challenging

Command Description

grep print lines matching a pattern

sed a stream editor; it allows you to change the contents of a file

and is often used in piped commands.
Options used:
-r Use extended regular expressions in a search
-e Add the script to commands to be executed
s/find/replace/g Change the string specified by find, by
string specified by replace. The g at the end stands for
global – on each line, all occurrences of the find string are
replaced. Without this option only the first occurrence
would be changed.

Version 4.2 (TT 2019) 38 IT Learning Centre

 Linux: a comprehensive introduction

cat print files on the standard output

diff compare files line by line, displaying the differences

Use the above to change all occurrences of nedles, needles, and neeedles to noodles.
Check that
grep -n -E 'ne{0,3}dles' HAYSTACK
finds all occurrences of nedles, needles and neeedles in HAYSTACK. Then enter
sed -r -e 's/ne{0,3}dles/noodles/g' HAYSTACK | cat -n
The '| cat -n' isn't necessary, but adds line numbers to the output to show which
lines have been changed.
Note that the regular expression used by sed is the same as the one used by grep.
This is a particularly useful feature of regular expressions.
Finally, send the output to a new file, NEW_HAYSTACK.
sed -r -e 's/ne{0,3}dles/noodles/g' HAYSTACK >NEW_HAYSTACK
All your nedles, needles and neeedles should now be noodles! You can compare the
two files with
diff HAYSTACK NEW_HAYSTACK
This completes the exercises on regular expressions. This is a very brief introduction:
they are a powerful and flexible tool.

IT Learning Centre 39 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

SHELL SCRIPT EXERCISES

Exercise 29 An example shell script
We're going to download a small file containing a bash shell script and change it.
Enter
wget https://www.stats.ox.ac.uk/pub/susan/linux/small.sh
in the terminal window. This command will copy the file small.sh from my website to
your Ubuntu desktop. Use the ls command to check this. You may need to make it
executable with
chmod +x small.sh
Now enter
./small.sh
and see what output appears. Now start editing the file with
gedit small.sh &
and change “Hello” to “Goodbye”, save the file and quit. Then run the file again to see
what happens.

Exercise 30 Developing your first shell script

You are now going to create your own version of the hello script. In the terminal
window type in the following command and an editor window should appear:
gedit hello &
• Type in the first version of the script – you need the two lines in the box above.

#!/bin/bash
echo "Hello world"

• Save the script - the file will be called “hello”

• In the terminal window enter the command
chmod +x hello
Remember how files have properties associated with them? One such property is
whether a file is executable. If a file is executable, you can run it (like a .exe file in
Windows).
chmod +x file
makes file executable by all users (hence a+x).
Now let's run the script. It is located in your current directory called . so we type:
./hello
You should see the output
Hello world
When you have a working version of the script copy your file “hello”to a new file
“hello1” so that if a later version stops working you can revert to a previously working
version.

Version 4.2 (TT 2019) 40 IT Learning Centre

 Linux: a comprehensive introduction

cp hello hello1

Version 2 of the script

#!/bin/bash
echo "Hello $1"

New concept: you can choose who to greet using a command line argument. This is
interpreted by the script with $1 .
• Now edit the file so that it looks like the version 2 of the script
• Run the script with a command line argument and check that it works
./hello Marge
You should see the output
Hello Marge
Again copy the “hello” file to a new file “hello2” to save a known working version.
Now repeat these steps for the four further examples:

Version 3 of the script

#!/bin/bash
for name in $*
do
echo "Hello $name"
done

New concepts:
• $* matches all the arguments entered on the command line
• the for name in list construct is particularly useful for operating on command
line arguments. The variable $name is assigned to each argument in turn and then
evaluated when used with echo .
After editing the file to make the above changes, run the command
./hello Bart Lisa Maggie
This should produce the output
Hello Bart
Hello Lisa
Hello Maggie

IT Learning Centre 41 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

Version 4 of the script

#!/bin/bash
if [ $# -eq 0 ] ; then
echo "No one to say hello to"
exit 1
else
for name in $*
do
echo "Hello $name"
done
fi

New concepts:
• $# contains the number of arguments supplied. You can check this by adding the
line
echo $#
above the line beginning if [ $# eq 0 ...
• Control flow – the idea that you can choose what happens next based on a condition
being met – is introduced using if expression then else fi.
• exit 1 causes the script to stop
After editing the file to make the above changes, run the command
./hello
You should see the output:
No one to say hello to
Now run the command for name in list
./hello Homer Marge Bart
which should produce the output
Hello Homer
Hello Marge
Hello Bart
as in version 3.

Version 4.2 (TT 2019) 42 IT Learning Centre

 Linux: a comprehensive introduction

Version 5 of the script

#!/bin/bash
if [ $# -eq 0 ] ; then
echo "No one to say hello to"
echo "Please enter people to greet"
read people
echo $people
else
people=$*
fi
for name in $people
do
echo "Hello $name"
done

New concepts:
• Prompting for input from the script. Instead of stopping when no command line
arguments are supplied the script now prompts for input. The name or names
supplied are stored in the variable people. At this point the names are stored in one
of two places so the else clause is used to move the contents of the command line
argument to the variable people.
• The for name in loop now uses people rather than $* for this list of names.
Make the above changes and run the command
./hello
now produces the output
No one to say hello to
Please enter people to greet
enter the names:
Homer Marge Bart
and hit return.
you should see
Hello Homer
Hello Marge
Hello Bart
Running
./hello Homer Marge Bart
will also produce

IT Learning Centre 43 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

Hello Homer
Hello Marge
Hello Bart
Now for you to try a couple of things.
• Create a test within the for loop to check for a particular name – say Homer – and
change the message for this name. Hint: the test might look something like this.
if [ $name == "Homer" ]
• Create a small file called Names containing a few names. Now run the following
command:
cat Names | xargs ./hello
Note how easy it is to change where the input comes from. You can download an even
longer list of names from here using
wget https://www.stats.ox.ac.uk/pub/susan/linux/Simpsons

Version 6 of the script

The final version of this script introduces a slightly more complicated concept.

#!/bin/bash
if [ $# -eq 0 ] ; then
echo "No one to say hello to"
echo "Please enter people to greet"
read people
else
people=$*
fi
for name in $people
do
echo "Hello `echo $name | rev`"
done

New concept:
The line beginning echo “Hello ... has been changed. Looking at this in more detail
shows that the there is `(backquote) character before echo $name and after rev. This
means that echo $name | rev is executed before being displayed on the screen. The
effect of this is to pipe the output from echo through the command rev, which simply
reverses the output.

The backquote is found at the top lefthand side of the keyboard next to 1.

Version 4.2 (TT 2019) 44 IT Learning Centre

 Linux: a comprehensive introduction

Running the command:

./hello Homer Marge Bart
will produce the output
Hello remoH
Hello egraM
Hello traB

that is the names have been reversed.

Exercise 31 File manipulation scripts

We are now going to look at some scripts to create and copy files. This script will
create several files with similar names.

#!/bin/bash
first=$1
last=$2
root=$3

while [ $first -le $last ]; do

touch $root$first
let first=$first+1
done

Save this in a file called create, give the file execute permission and then use
./create 1 4 newfile
which would create four files called
newfile1 newfile2 newfile3 newfile4
A while loop is used to repeatedly create files until the number specified is reached.
The line let first=$first+1 adds one to first each time the loop is executed.
Change this script to check that the right number of command line arguments have
been given. Print out a message and stop if there aren't three arguments.
The answer is in the Answers section.

Now we will look at a script to copy all files with the same root to another name.

IT Learning Centre 45 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

#!/bin/bash
let n=1
for file in $1*; do
cp $1$n $2$n
let n=$n+1
done

So assuming the script is in a file called mycopy

./mycopy newfile test
All files called newfile1 ... newfilen will be copied to test1 ... testn.
Try it. What does the $1* in the line beginning for file ... do? Change the script so that
the new file name has a dot between the name and the number. [This is very easy!]

[More difficult.] Change the script so that the new file name has a dot after the
number and then the date of the form 03Mar13 so that the files look like
test1.27Feb13 ...
You will need to look at the date command – use date --help or man date and
Version 6 of the hello script. The answer is below.

Note that there is a very useful command rename which will rename one or more files
based on a regular expression. It may not be installed by default. If you want to
investigate this enter
sudo apt-get install rename
rename --help
apt-get install is the command line tool for installing extra command or packages.
There is a brief discussion of sudo in the next session.

Concluding Remarks:
These exercises can only scratch the surface of what can be done with the command
line and shell scripts. Don't expect everything you've seen today to sink in. Working
with command lines takes some getting used to and the best way to learn is by lots of
doing. For now, simply take away the ideas of what this environment is capable of.

THIS COMPLETES THE EXERCISES FOR SESSION THREE

Version 4.2 (TT 2019) 46 IT Learning Centre

 Linux: a comprehensive introduction

5 Using remote computers

Sometimes you will need to access a remote computer. This may be from home to a
Department system, or from a desktop computer to a more powerful computer.

Exercise 32 Setting up access to IT Services Linux system

First you need to set up access to the IT services Linux system.
Go to https://www.it.ox.ac.uk/ and click on ‘Manage Accounts’ on the left side of the
bar at the top of the page. You will need to use your Single Sign-On (SSO) account to
sign on.
You should see a page displaying accounts you can register for and services you can
use. In the ‘Other facilities available’ section, you should see ‘Manage linux shell
account’.

Click on this link and you should now see the screen to activate the account.

IT Learning Centre 47 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

If you are asked to choose a shell, make sure you choose /bin/bash. You should then

Exercise 33 Using ssh

Now that the account has been enabled, you can log on.

Command Description

ssh ssh connects to a different Linux/Unix system.

Use
ssh [email protected]
at the login prompt enter your SSO and password. So if your SSO is coll1234 you
would type in
ssh [email protected]
You should see a screen like this:

Version 4.2 (TT 2019) 48 IT Learning Centre

 Linux: a comprehensive introduction

The commands that you have used on the live Ubuntu, will also work on this system,
although it is running a different Linux distribution. It is running Debian Jessie.
Use the cd, wget and tar commands from Exercise 14 to download the files to this
server.
cd
wget https://www.stats.ox.ac.uk/pub/susan/linux/LinuxFiles.tgz
tar -xvzf LinuxFiles.tgz
Make sure that commands are behaving in the same way by using examples from
previous exercises.

Command Description

ssh ssh connects to a different Linux/Unix system.

Options used:
-X Allow X11 forwarding.

It is possible – and sometimes necessary – to open a graphical application on the

remote system. To do this logout using
exit
and log back in again from your local Ubuntu desktop with
ssh -X [email protected]

On linux.ox.ac.uk enter

xclock &

You should see a window like this appear:

Exercise 34 Copying files between systems

As well as logging on to remote systems, it is often necessary to copy files and
directories from one system to another.

IT Learning Centre 49 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

Command Description

scp Copy files and directories between systems.

You will need two terminal windows open.

• The local live Ubuntu
• A remote session on linux.ox.ac.uk
On linux.ox.ac.uk, create a new directory to store the files and directories you will be
copying.
cd
mkdir Copies
Use ls Copies to make sure it is empty.
Now in the local live Ubuntu terminal type in
scp HAYSTACK [email protected]:~/Copies/.
Make sure you type the command exactly as it appears.
The command is quite complex so each component is explained in detail here:

scp command Description

scp Copy files and directories between systems.

HAYSTACK Name of file to be copied.

SSO Username on remote system. Replace by your single

sign-on.

@ Location indicator.

linux.ox.ac.uk Name of remote system

: Separator between remote system name and location

of file.

~ A short cut which refers to your home directory.

Useful if you don’t know the path to your home
directory.

/Copies/ The directory where the file will be stored

. “.” indicates that the file will be given the same

name as the original.

Now in the linux.ox.ac.uk terminal use

ls Copies
to make sure the file has been copied.

Version 4.2 (TT 2019) 50 IT Learning Centre

 Linux: a comprehensive introduction

See if you can copy HAYSTACK from the live Ubuntu session to the same location but
with a different name.
See if you can use wild cards to copy all the files beginning foo2 from the Wildcards
directory to the same location and with the same name.
Use ls Copies on linux.ox.ac.uk to make sure the files have copied successfully.

In both cases, see the Answers page for a suggested solution.

Finally, remember that wget will download datasets available from the web.

Exercise 35 Copying directories between systems

scp command Description

scp Copy files and directories between systems.

Options used:
-r recursively copy the contents of a directory

Files Name of directory to be copied

On the local live Ubuntu terminal enter

cd
scp -r Files [email protected]:~/Copies/.
On the linux.ox.ac.uk terminal check that the directory and contents have been copied.

Exercise 36 Managing sessions on remote systems

Command Description

screen Connect and disconnect from a session, possibly from

multiple locations, and allow long-running processes to
persist without an active shell session.
Options used:
-r reattach a screen session
-list display all screen sessions

In the linux.ox.ac.uk terminal use

screen
to start the screen session. You should see a window like this:

IT Learning Centre 51 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

Press return and you should see the standard prompt. Now use wget to get the script
you will be using
cd
wget https://www.stats.ox.ac.uk/pub/susan/linux/my_date.sh
Permanently change the permissions, so that the script can be executed.
chmod +x ./my_date.sh
Have a look at the my_date.sh script to see what it does. Don’t worry if you can’t
understand every line.
cat my_date.sh
#!/bin/bash
# Ask for name.
# Repeatedly output the name and date, waiting 15 seconds.
# Terminate after 15 minutes.
echo "Please type in your name"
read name
start=`date +%s`
now=$start
while (( (( $start + 900 )) > $now )) ;
do
echo "Hello $name, the time now is `date`"
sleep 15
(( now += 15 ))
done

Version 4.2 (TT 2019) 52 IT Learning Centre

 Linux: a comprehensive introduction

The script displays the date and time every 15 seconds for 15 minutes. Run the script
interactively to watch what happens.
./my_date.sh
Watch the output appearing for a minute or so. You should see something like this:

Now we are going to detach from the screen session so that the script will continue
running but we will logout of linux.ox.ac.uk. To detach from the screen session use
^CTRL-a d
That is, hold down CTRL and a, then press D. You should see a message
[detached from XXXX.pts-XX.raven]
where each X is a digit.
You can check what screens you have with
screen -list
and should see something like this:
There is a screen on:
6144.pts-30.raven (01/23/17 17:29:58) (Detached)
1 Socket in /var/run/screen/S-hutchins.

Logout from linux.ox.ac.uk

exit
Perhaps from another desktop (swap with your neighbour!), log on again to
linux.ox.ac.uk
ssh [email protected]
replacing SSO by your single sign-on.
Reattach the screen
screen -r

IT Learning Centre 53 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

and you should see the list of times with no interruption.

This is a slightly contrived example. In general output from long running jobs is best
saved in a file rather than displayed on the screen, as there limits to the amount of
data that a terminal displays. However, the technique for detaching the screen and
logging out is the same.

Concluding Remarks:
Accessing remote computers is a common requirement for Linux users. On Windows, a
small application called PuTTY provides a secure means for logging on to a Linux
system. On Macs, use the terminal window.

Version 4.2 (TT 2019) 54 IT Learning Centre

 Linux: a comprehensive introduction

6 Answers
Session 1
Exercise 3 Try Dock | Files
Exercise 4 The active application is highlighted in a lighter coloured square. For
each open application an orange dot appears.
Exercise 4 Keyboard shortcuts
To create keyboard shortcuts to switch between desktops
1. Open System Settings | Keyboard
2. Click on Navigation
3. Scroll down until you find Switch to workspace 1
4. Click on Disabled
5. Hold down the Ctrl key and press F1 (on the top lefthand side of the
keyboard)
6. Now right click on Switch to workspace 2
7. Set this to Ctrl and F2
8. Do this for workspace 3 and workspace 4 ...
9. Click on X to save the changes.
If you now hold down the Ctrl key and press F2 you should switch from
Desktop 1 to Desktop 2

Exercise 13 You can also create files in /tmp.

Session 2
Exercise 16 Finding a drawing application
This is straightforward:
apropros draw
lists several possible matches. The most useful is
lodraw (1) - OpenOffice.org office suite
If you enter lodraw & on the command line the application will start.
Exercise 17 Reading a file called *star
cat \*star
cat “*star”
are both possibilities. A \ (backslash) is an escape character which stops the star
being interpreted as a wildcard. Surrounding file names that contain special
characters with quotation marks often work. There may well be other ways to do this.

IT Learning Centre 55 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

Changing to a directory called Open This.

cd ..
cd Open\ This
but again there will be alternatives.
Reading a file called -ReadMe. Again there are lots of possibilities
cat ./-ReadMe
cat -- -ReadMe
both work. In the first suggestion (cat ../-ReadMe) using a relative path overcomes
the problem of the leading dash; in the second (cat -- -ReadMe) the cat command is
given an extra option of two dashes which alerts it to the fact that the filename starts
with a dash. Both these options work with the more and less commands too. Similarly
to delete the file use
rm -- -DeleteMe
rm ./-DeleteMe

Excercise 19 Wildcards
ls foo? matches
foo1 foo2
ls foo2* matches
foo2 foo20 foo2bar
ls foo[1-2] matches
foo1 foo2
To match just the files foo20 and foo2bar use
ls foo2?*
To use wc -l just to display the files with a 1 in their name.
wc -l *1*

Exercise 20 Searching and sorting

Use grep -i melon file to find all occurrences of melon whatever the case.
Use sort -r fruit veg to reverse the order of the sort.

Exercise 22 Finding the largest file

Use du -sk /usr/bin/s* | sort -n | tail -5 to find the 5 largest files beginning
with 's'.

Exercise 23 Merge information from different files

awk '{print "A", $2, "eats"}' creatures | paste -d" " - veg

Version 4.2 (TT 2019) 56 IT Learning Centre

 Linux: a comprehensive introduction

Session 3
Exercise 26 Simple regular expressions
grep -n -E ^needles$ HAYSTACK
will find all lines with just one occurrence of the word needles with no white space
either before or after. We'll look at how to match white space in the next set of
exercises.

Exercise 27 Regular expressions with spaces

No matches are found, because this search is looking for matches where needles has
characters both before and after it - hayneedleshay for example - and there are no
instances of this in HAYSTACK.

Exercise 32 File manipulation scripts

cp $1$n $2$n.`date +%d%b%y`

Session 4
Exercise 35 Copying a single file between systems
scp HAYSTACK [email protected]:~/Copies/HAYSTACK2
scp foo2* [email protected]:~/Copies/.

IT Learning Centre 57 Version 4.2 (TT 2019)

Linux: A comprehensive introduction

Copyright

Susan Hutchinson makes this document and the accompanying LibreOffice Impress
presentation available under a Creative Commons licence: Attribution,
Non-Commercial, No Derivatives. Individual resources are subject to their own
licencing conditions as listed.
The Oxford University logo and crest is copyright of Oxford University and may only
be used by Oxford University members in accordance with the University’s branding
guidelines.
Acknowledgements
This course was originally developed with Jon Lockley.

Version 4.2 (TT 2019) 58 IT Learning Centre

 Linux: a comprehensive introduction

Version information
Minor updates – replace Usenet with StackExchange May 2014 SRH
Updated for Ubuntu 14.04 (Trusty Tahr) Sep 2014 SRH
Minor corrections Jan 2015 SRH
Minor corrections and checking. Regular expressions. Nov 2015 SRH
Major changes to Section 4: replaced software installation with remote Apr 2016 SRH
access to linux.ox.ac.uk. SRH
Jan 2017
Added branded front page Apr 2017 SRH
Corrections for 2017-2018 Sep 2017 SRH
First section rewritten for Bionic Beaver, 18.04 Oct 2018 SRH
Typos fixed immediately after MT2018 presentation
Updated footer for HT 2019 Jan 2019 SRH
Minor corrections TT 2019 Apr 2019 SRH

IT Learning Centre 59 Version 4.2 (TT 2019)

 Computer platforms: Your safety is important
Linux: a comprehensive
introduction
Susan Hutchinson
[email protected]
• Where is the fire exit?
• Beware of hazards:
Tripping over bags and coats
• Please report any equipment faults to us
• Let us know if you have any other
concerns

MT 2018
 Your comfort is important

Session 1
• The toilets are along the corridor outside Introducing the
the lecture rooms.
• The rest area has vending machines and
Linux desktop
a water cooler.
• The seats at the computers are
adjustable.
• You can adjust the monitors for height,
tilt and brightness.
 IT Services Linux Courses Session 1

Today's course is divided into four parts each ●

What is Linux?
consisting of:
– Open source and why it is important.
●
A short presentation.
●
– Where it came from and how it is made.
Exercises.
●
The four parts are What is Linux used for?
● ●
Introducing the Linux desktop. How to get Linux.
●
Simple use of the command line. ●
Linux office applications.
●
Further use of the command line and shell ●
scripting.
Your questions and a practical session.
●
Using remote computers and managing your
own computer.

If you are comfortable using office applications, browsing the web and using
email then this course aims to show that you can use Linux without too steep a
learning curve. The first two sessions focus on the similarities between Linux
and other operating systems such as Windows and Mac OS.

In the remaining sessions we start to explore the differences between Linux and
Microsoft Windows. The course demonstrates many powerful features that
have no equivalent in Windows.
 How Linux changed things What is Linux?

●
Linux is an operating system designed by Linus Torvalds in the early
Linux does not come from a single large corporation. 1990s.
●
It offers an alternative approach. An operating system protects the user from the hardware and vice versa.
– Freedom of choice. An operating system has several parts:
– Freedom to understand and change. – kernel
– Software written with quality rather than profit as a – system calls
goal. – programs
●
Competition and alternative approaches benefit users Other Operating Systems include
and consumers.
– Windows (7, Vista, XP, 2000, 95)
– Software developments: firefox for example.
– MacOS
– Increased awareness of open standards by Microsoft.
– Unix
– Easier to work using different systems.

The Operating System manages resources such as the processor, disks and
Let's take the example of browsers. Internet Explorer is now quite a good memory, and provides a consistent interface through system calls for
browser, but it needed Firefox – and now Google Chrome – to make this application programs. This means that application developers do not need to
happen. For example, would tabbed browsing have been added to IE know details of the hardware.
without Firefox? Furthermore, issues around security in browsers have
been hugely improved. In this case competition has improved the An amusing analogy to give you a flavour of different operating systems is “If
experience for everyone. Operating Systems Ran Airlines”. Enter this into a search engine to find out
more.

Linux is related to Unix and certainly appears very similar but works rather
differently underneath.

There are other free Unix operating systems if Linux doesn't appeal: see
www.freebsd.org for example.
 Who makes Linux? How is Linux developed?

Linux is maintained and developed both by businesses and

Linus Torvalds created the first Linux kernel in August private individuals.
1991:
●
Hundreds are involved in work on the Linux kernel.
“Hello everybody out there using minix- ●
I'm doing a (free) operating system Thousands work on the applications which come with Linux
(just a hobby, won't be big and distributions.
professional like gnu) for 386 (486) AT It is known as open source software.
clones. This has been brewingsince
april, and is starting to get ready. ” Linux or GNU/Linux?
●
Text of original email from Linus to the comp.os.unix Linux is the kernel.
newsgroup sent on 25 August 1991. ●
GNU is the project to create a Unix-like operating system.
GNU stands for GNU's Not Unix (terrible Unix joke).

Linus Torvalds developed the first versions of Linux because he wanted to

run a Unix-like system on his home PC. This version of Linux did little more A comprehensive description of the case for open source software and the
than allow Linus to log onto his university's system to check his email. problems with closed source software can be found in the book The Cathedral
However the project quickly grew into a full operating system kernel and and the Bazaar by Eric S. Raymond, O'Reilly (2001). The book is also
Linus published it as an open source project. available online. Raymond is a well-known open source advocate so some
sections of the book reflect his strongly held views but as a whole the book
A good book about the early history of Linux is Just for fun by Linus does make a good case for open source.
Torvalds and David Diamond, pub. Harper Business (2001).
Alternatively www.fsfeurope.org will tell you more about free software.

There is more information about GNU, its relationship to the Linux kernel and
the GPL (General Public Licence) at www.gnu.org .

GPL is the standard licence for most releases of free, open source software.
Several different notions of freedom are incorporated in the idea – freedom to
change, freedom to understand, freedom to distribute.

One very important point to remember is that all Unix jokes are truly awful.
 What is open source? What is Linux used for?

while(fscanf(fd, "%f", &y)

==1){ 11001010110110 Linux was originally designed for desktop use but this
n++; 10010011101010
sum += y;} Compiler did not make a significant impact on the market.
if (n > 0) 10010000110101
(void) printf("The MEAN 01011010010100 Linux did quickly gain a large following for
is %f\n", sum/n)
11010101001011 ●
Servers – for example, web servers
The source code written by the programmer The binary (or executable) which you run
●
Internet services (DNS, routers etc)
Open Source means:
Closed Source means:
●
 Programming
We can see binary and source. 
We are only given binaries.
 ●
We are encouraged to read and 
We cannot read source. High performance computing
understand source code.

We cannot recompile code.
 Significantly improved user interface and economic
We can debug and recompile
source. advantages mean that Linux is gaining popularity on the
desktop.

In Windows .exe files contain machine-ready instructions. In Linux [and Much of the University infrastructure, particularly at IT services, is based on
Unix] these sorts of files are called binaries (although they are usually machines running Linux. IT services organise a tour of the machine room
executable). This is because there are also human readable executables every year and you'll find a penguin badge on most things.
or “scripts” which are lists of commands. Further Linux [2] looks at scripts.
The web also relies heavily on Linux – many web servers are Linux based
What are the advantages of building software from source? When you are machines.
provided with only the binary you can be sure that it wasn't built on your
machine. Building your own binary from source means that you can specify Linux desktop systems were very poor compared with commercial
that it should be optimised for your own hardware and customised with alternatives until a few years ago. This made it unlikely to appeal to non-
features that are or are not needed. technical users who were using MacOS or Windows. Linux also lacked any
good office software. Both of these areas – as we hope to demonstrate –
If something goes wrong you could debug the software – or at least get have now been addressed very successfully. However this has left a legacy:
someone else to do it for you. You can also change the way it works if it Linux still has the reputation of being hard to use.
only nearly does what you need.

But you don't have to!

 Getting Linux Linux distros

The easiest way to get Linux is to download or to buy a “distro” or The two big distros are RedHat/Fedora and Suse:
distribution.
A distribution = a kernel + applications + installation software+
support + documentation.
Typical applications are
●
Office software
Debian Mint
●
Databases
●
Programming languages and tools
● Mepis Slackware
Web browsers
●
Email readers
●
Internet services (www, DNS, NIS, firewalls) Mageia arch
●
Games

Most distros come with a version number – for example Fedora 28 or There is no technical reason why a RedHat, Fedora or openSuse (now
OpenSuse 18.04 but they are not particularly significant. When acquiring owned by Novell) installation cannot be done at home. However, in recent
Linux the important thing is to find out what version of the kernel is supplied years both distros have changed their product lines to focus more on the
and what versions of software are included. It is simple – with Internet business user. RedHat are also involved with a free distro called Fedora
access – to upgrade to the latest version but it is good to start with relatively which is now stable. There is also another distro, CentOS, which is based on
recent versions. RedHat.

Current kernel versions are 3.16.x, 4.14.x, 4.18.x Of the other free distributions Debian is the most mature and stable but is not
recommended for the newbie (beginner in Linux-speak). Debian variants like
● Ubuntu are more friendly. Mandriva, Mint and particularly Ubuntu are all
Office software might be LibreOffice or Calligra
● aimed at the home user and are a good place to start if you want to try linux
Databases might include mySQL
● at home.
Programming languages should include C, C++, Fortran, Perl and Python
●
Web browsers might include Opera, Mozilla Firefox, Midori and
Chromium Free Linux distros can be downloaded from www.linuxiso.org .
●
Email readers might include Mozilla Thunderbird, Kontact, Gnome Gmail
See http://distrowatch.com/ for a discussion on recent Linux developments.
Generally, Firefox and Thunderbird, the browser and mail reader from the
Mozilla stable are included. Both of these are also available from
www.mozilla.org and have versions for Windows as well as Linux.
 Ubuntu Linux and Windows

For this course we will be using the Ubuntu distro ●

You can install Linux on your PC without removing
Windows (or any other operating system).
This version of Ubuntu is a special kind of Linux known as a – At boot, choose which to use.
“Live” system and has two significant features:
– Could use emulation software (VMWare, wine).
●
It runs from USB.
●
●
Can share devices using samba.
The hard drive is not touched.
●
It's safe to run this on a Windows system without installing or
Linux will read most Windows file systems.
changing anything. ●
Windows won't support Linux file systems.
Ubuntu is a Debian-based version of Linux, providing a
simple desktop and straightforward installation. We are using
Ubuntu 18.04 LTS (Long Term Support, Bionic Beaver)

We are using Live Ubuntu for this course because it does not change or use
the hard drive and so IT Services does not have to rebuild machine Because Windows is still the dominant desktop system Linux works hard to
especially for this course. This also means that you can try this version of support Microsoft standards so that Linux and Microsoft users can work
Ubuntu at home safe in the knowledge that is very, very difficult to damage together.
your machine or the data you have on your disk.
VMware creates a virtual machine which can run binaries for the operating
The disadvantage of live systems is that it is more fiddly to save data system it emulates. You can run a virtual Windows machine under Linux or
permanently because the hard drive is not easily accessible. However some the other way round. VMware is not free or open source.
people find live Ubuntu and a USB memory stick a very useful combination.
The slides and exercises were prepared like this. Often, you will need to Wine is an open source copy of the Windows system libraries which allow
install software as there is only a limited range on the USB. many Windows .exe files to run under Linux. Many software packages run
faster under Linux+Wine than under Windows!
If you decide that you like Linux it is better to install a more permanent
version on your hard drive. WineX is a more powerful but not free version

To download Ubuntu go to http://www.ubuntu.com. We are using Ubuntu Oracle (formerly Sun) provide an open source product, VirtualBox which is
18.04 LTS (Bionic Beaver). LTS versions will continue to get updates for 4 becoming increasingly popular.
years after release, other releases need upgrading after about 1 year.
 Desktop Environments

●
Desktop systems in Linux come with a variety of
desktop environments.
Office applications ●
Provide an easy to use graphical use interface
or GUI based on the X window system.
●
Choice: KDE or Gnome are the most widely
available.
●
Generally they provide very similar functionality.
●
Competition has produced two very usable
solutions.

If you are comfortable using office applications, browsing the web and
using email then this course aims to show that you can use Linux
without too steep a learning curve. The first session starts by focusing
on the similarities between Linux and other operating systems such
as Windows and Mac OS.

In the remaining sessions we start to explore the differences between

Linux and Microsoft Windows. The course demonstrates many
powerful features that have no equivalent in Windows.

Unix systems have had a graphical user interface for some time but
these have been largely proprietary. Clearly the GNU/Linux project
needed an open source desktop environment that would provide an
intuitive user interface.

A graphical user interface comprises two components: the X window

system which is very loosely analogous to the kernel and a choice of
desktop environments which provide the user with the familiar
windows, menus, backgrounds and so on.
 LibreOffice and Microsoft #1 LibreOffice and Microsoft #2

●
LibreOffice runs on Windows just as well as on Linux Microsoft Office products and their LibreOffice
and Unix. equivalents:
●
LibreOffice supports all Microsoft data formats. Microsoft Word = Writer
●
LibreOffice offers almost all the functionality that
Microsoft office does (a few fancy bits might be
Microsoft Excel = Calc
missing). Microsoft PowerPoint = Impress
●
LibreOffice was far more standardised than Microsoft Microsoft Paint = Draw
Office but the latest version of Microsoft Office does
allow you to save files in an open standard (XML)
format.

Most Unix users used (and Mathematicians still use) TeX and LaTeX (the X is
pronounced to rhyme with “ch” as in loch) to produce documents. These are very
All of these applications are installed in Live Ubuntu. The word processor,
powerful typesetting tools and give far more control than word processors. They are
spreadsheet and presentation software can be found are all available from
excellent for writing theses, books and papers, particularly those with formulae and
the Dash and the draw application can be found by clicking on the Dash
pictures but not suitable for short documents and letters. They also have a slide-
icon and searching for Draw.
making environment which is fairly simple to use simply but can be rather difficult to
use for complex presentations.

There were other typesetting programs such as nroff/groff/troff which were used for
creating manual pages. As you may now predict there are a range of packages
available which provide office productivity tools. These usually include a document
creator (or word processor), a spreadsheet, a tool for creating presentations and a
drawing package.
Microsoft files do not always read perfectly in LibreOffice. Microsoft – as is their right –
do not reveal how data in MS files is formatted and these formats are changed from
time to time. This means that LibreOffice software developers have to respond to
these changes and “reverse engineer” files in order to code software to read files.
Clearly not sharing document formats is a good way to make the development of
other office software more difficult...

Generally I find that LibreOffice files are substantially smaller than Microsoft files.

More recently Microsoft have started to use some open standards and it is possible to
save Microsoft documents in XML.
 Getting help The command line

Getting help with Linux questions can be difficult Why use the command line?
so it helps to know where to look. The sites I use ●
How does the command line work?
often are: ●
What is a shell?
●
StackExchange – a network of community- ●
Some simple commands.
managed sites providing expert answers to ●
Files and directories.
questions. ●
Hints and tips for file names.
●
Google – particularly when I have an error ●
Getting help.
message I don't understand.

StackExchange provides a useful source of source of support. It is a

gateway to a network sites answers to questions. Anyone can ask a
question and anyone can answer the question, which are are ranked and
rated. All sites are moderated so that the content is generally reliable.One
useful feature quoted on the 'About' page is: “Stack Exchange is not the
place for conversation, opinions or socializing”.
 How commands work
The command line
A command at its simplest is of the form:
●
A graphical user interface (GUI) is available in both command
Windows and Linux.
Often commands require an argument such as a file name
● command file
The command line is often unfamiliar to Windows
users.
Some examples are
●
Compare Windows and Linux when reading a PDF date [to display the time and date]
file or starting a browser. gedit newfile [start the gedit editor to
●
Either double click on the icon or enter the change the file newfile]
command
evince file.pdf &
firefox &
This is the format of many commands.

It is also possible to modify the behaviour of a command using options. So

We have now spent some time looking at the Ubuntu Graphical User for example the command ls on its own will display the name of all the files
Interface. We will now focus on the command line. This can seem an in a directory.
unfamiliar place for Windows users.
ls
For example when reading a PDF file in Windows you will either click on
the picture or open the application from the start menu and browse to the By including options you can display details about the size, last changed
file. In Linux you are just as likely to enter date and access permissions of a file and many many other details. Options
evince filename are specified by a hyphen and a single character so
on the command line. Similarly the mozilla browser can be started by ls -l
entering
firefox would give a long listing. This is something of a simplification but will do for
When you execute a command that opens a new window it is customary to now.
include the ampersand character - & - before pressing enter. This allows
you to continue using the terminal window. So
firefox &
The & is only necessary when the command opens a new window; you
don't need it for other commands.
 Speeding things up Upper and lower case

Surely all that typing can't be right! How can we Case sensitivity
speed things up? ●
Linux commands and filenames are case sensitive:
Filename and command completion: BIG is different from big.
●
<tab> key completes commands and filenames ●
Almost all commands use lower case.
●
In Windows case is not significant. If you save a file as
What else can we do to save typing? big.doc, type in BIG.DOC and Word will find it.
● ●
arrow keys allow us to: However, if you are working with files in both Linux and
Windows you need to take care: Windows will see BIG.DOC
– recall previous commands and big.doc but can only use BIG.DOC.
– change previous commands Guidelines for file names: don't include:
● ●
spaces ● * and ? ● hyphens / and \

The shell is the command that interprets the commands we enter. As you
might expect there are many different shells. We will be using the bash shell.
One significant difference between Windows and Linux is that Linux is case
There are many shortcuts that allow you to use the command line more sensitive. This means that a files called
efficiently. For example, the tab key can be used to complete both commands big
and filenames. If you enter the command Big
BIG
ev are all different files. In Windows this is not so. This difference is important if
you are sharing information between Windows and Linux.
and then press <tab> once you should hear a beep. That's because there is
more than one command beginning with ev. Pressing <tab> twice in quick As we have seen Linux commands use a space as a separator between
succession should display a list of commands beginning with ev . If you then commands and files. This means that spaces in file names can cause
enter the letter i so that you have evi and press <tab> again the command problems. How would you read a file called My File? For similar reasons,
evince should appear. hyphens (-), slashes (/) and backslashes (\) are best avoided in filenames.
 Two things Linux does better Exercises for Session 1

... in my opinion! The aims of these exercises are:

Multiple desktops ●
To look at the desktop.
● ●
reduce clutter To explore some components of LibreOffice.
● ●
organise work logically To take a first look at the command line.
Flexibility ●
Please feel free to:
● ●
better organised desktop Ignore desktop exercises which are not relevant
● your work.
powerful command line
●
● Try to do the tasks in LibreOffice that you do
choice of many applications
with Microsoft Office.

Most distros come with a version number – for example Fedora 18 or

Ubuntu 12.04 but they are not particularly significant. When acquiring Linux You can download LibreOffice free from
the important thing is to find out what version of the kernel is supplied and www.libreoffice.org/download/ .
what versions of software are included. It is simple – with Internet access –
to upgrade to the latest version but it is good to start with relatively recent OpenOffice is available from www.openoffice.org.
versions.

Current kernel versions are 2.6.x and 3.5.x.

Office software might be LibreOffice or Koffice.

Databases might include mySQL.
Programming languages should include C, C++, Fortran, Perl and Python.
Web browsers might include Opera, Mozilla, Netscape and Konqueror.
Email readers might include mutt, pine and evolution.

Also look out for Firefox and Thunderbird, the browser and mail reader from
the Mozilla stable. Both of these are available from www.mozilla.org and
have versions for Windows as well as Linux. Finally the google chrome
browser is also widely used.
 Session 2

We're now going to look in more detail at

Session 2
the shell and the command line
Using the ●
What is the shell?
command line
●
Navigation – how to move around.
●
More advanced use of commands.
●
Pipes or how to build your own
commands.

Ask for any questions.

F
 Unix/Linux commands
Shells
We can interact directly with a shell with a shell prompt aka
the Command Line Interface
A shell sits between the user and the kernel.
The UNIX Philosophy: commands should
There are several ways of using shells:
●
do only one small task;
– Graphical desktops (indirectly). ●
do it quickly;
– Running applications (indirectly). ●
do it quietly and don't ask for confirmation.
– Command line interface (directly). If a new requirement arises start again - don't add features to
an existing command.
– Scripts (directly).
Advantages and disadvantages with this approach:
●
● Many commands so each is small and simple.
There are different shells available for Linux
●
●
A lot to remember.
The default shell in Linux is called BASH.

You may not realise it, but when you are using the graphical desktop that it
is merely a collection of applications running on top of a shell. Most of the commands available on early Unix systems are still used
today, as well, of course, as many new ones that have since been added.
What we are going to look at today are ways of interacting with shells in a
more direct manner. Unix commands tend to be small and have a single function. This means
that there are a lot of commands to remember. Another side effect of the
long history is that there is generally more than one way to achieve the
end you want. Often there is no correct solution although some are neater
than others.

It might also seem that giving commands a deliberately obscure name is

part of the philosophy. I'm not so sure about that - generally there is an
explanation for the name of a command although it isn't always
immediately apparent. Believe it or not there is a sensible explanation for
why the commands cat, grep and awk are so called!
 A digression on paths
Files and directories
Paths describe the location of a file in the filesystem.
There are three sorts of paths:
We can use commands to explore the way Linux
●
locates and stores files. simple a file or directory name
A Linux system looks a bit like a tree. Here's and newfile
example of the top (or bottom) layer: ●
relative a reference to a file or directory from the
current directory
../tmp/newfile

●
absolute a reference that will work anywhere
/home/ubuntu/Desktop/tmp/newfile

Linux, like Windows, organises files in a hierarchical directory structure.

There are some minor differences: the directory separator in Linux is /
Linux, like Windows, stores files in a hierarchical way with files stored within whereas in Windows it is \.
folders (except in Unix and Linux we call them directories). You can also have
directories within directories. Files can be reference in three ways. If, for example, there is a file called
myfile in your home directory
In Linux and Unix, everything starts at a single place called “/” or sometime,
the root. Some people think of root as the top of the structure, some as the /export/home/student/username
bottom. This is unlike windows where there might be a C:, a D: and an E: for
example. you can use more to view it in three different ways:

In the above diagram you can see that / is itself a directory (it has to be to more myfile
hold the rest of the file system) and contains a number of directories. I've
marked those typically belonging to the “system” in red and those you often more ./myfile
store files in in blue.
more /export/home/student/username/myfile

The first is a simple path, the second a relative path and the third a direct
path. In the first case you need to be in the same directory to see the file,
in the second you indicate the location in terms of where you currently are
and in the final case you will be able to see the file from wherever you are.
Commands and navigation Finally … some commands

The file browser moves you around

the system and manipulates your files. Orientation and navigation or where am I and where
Can use commands instead
do I want to go?
cd = change directory
●
pwd [print working directory]
ls = list files
●
cp = copy cd directory [change directory]
mv = move ●
/ cd [go to your home directory]
Learn to navigate up
pwd = where am I? and down directories ●
home cd .. [go up one directory]
mkdir = make new directory
user1 user2 user3
rm = remove file
Desktop Work Fun
rmdir = remove directory
datafile.txt

You'll probably be used to navigating though a file system with a You can explore the directory hierarchy using
graphical interface like Internet Explorer. The File Browser serves a
similar purpose. The difference is that IE is regarded as an integral cd ..
part of the Windows OS. The File Browser on the other hand is “just
another application” which you can easily remove and do without. to move up to the next level and
The dependence of Windows on big, complex tools like IE is one of
its major weaknesses. cd

Where UNIX relies on tools (commands) to do things, it demands to return to your home directory. Your home directory is where files
they are small and simple. you create are stored and where you start each new session. Your
home directory will be something like
Note the typically terse names of the Linux commands. They are
usually short for something (pwd is short for print working directory /homes/username
for example). or
/home/ubuntu
… and more ... … and more

File and directory manipulation

There are several ways of looking at the contents of a file:
●
ls list files in a directory
cat file look at the contents of one or more files.
– ls -a list all files in a directory
more file look at the contents of a file at the end of
– ls -l display a long listing each screenful.
– ls -la display a long listing of all files
less file like more only with more options.
●
mkdir directory create a directory
Some other commands
●
rmdir directory delete a directory
du -sk file find out the size of a file
●
cp file1 file2 copy file1 to file2
●
Sort sort the contents of a file, line by line
rm file delete a file
●
mv file1 file2 move file1 to file2

The cat command is said to be named after the verb concatenate as it can
also be used to concatenate several files into one. The command displays the
contents of a file on the screen. It is only really useful for small files.

more displays a file, pausing at the end of each screenful.

<space> moves to the next page
q quits at any time
<return> moves on a line

less has all the above options and several more. One useful one is
G move to the end of the file
 Other commonly used commands What is a file?

man get help with a command In Windows, files:

whatis short description of a command ●
Usually include formatting and application
lp print a file information.
●
Some more powerful commands Are used by only one application.
●
find find files matching pattern
Have a suffix (part of the name after the .) that
determines what sort of file it is.
grep search for a pattern in a file
wc report the number of characters/words/lines In Linux, files:
●
find . -mtime -5 -exec grep -i statistics {} \; - Are often plain text.
ls ●
May be manipulated by many applications.
We will be looking at more complicated commands next session. ●
Don't need a suffix to determine their type.

The man command is particularly useful for finding out all the available options
in a file. So A file in Windows is usually (but not always) closely linked to an
application. So a .doc file is considered to be a Word document, a .xls file
man ls an Excel spreadsheet and so on. Although we cannot see it, a lot of
information about the file and the application that runs it is included in the
will give you a vast amount of information on ls including the options. If you file.
are not sure of the name of a command but know what it does then you can
do a keyword search with the apropos command. If you can't remember In Linux, files often contain only plain text. When you look at the contents
what a command does then use of the file you see all there is. Files are not linked so closely to one
application: a file can be viewed, changed and compiled or printed by
whatis ls several different programs.

The commands find and grep are particularly useful. find can search
down through directories looking for files matching a given condition such as
all files changed within the last 7 days. grep will search a given file for a
particular pattern. The two commands are often used in conjunction like this

find . -mtime -7 -exec grep -i hello {}\; -ls

It has to be said that the syntax is obscure but this command would search all
files changed in the last 7 days containing the string “hello”.
 Wild cards and globbing Standard input and output

This slide could be called file name expansion! Linux commands make use of the concepts of
The most commonly used special characters to represent
parts of a filename are: Standard Output (STDOUT)

* matches none or more characters

? matches a single character
Command
[ ] matches any characters in a given range
Standard Input (STDIN)
[! ] matches any characters not in a given range
So to list all files whose name ends with .txt use Command
ls *.txt

The use of wildcards is another way that the shell makes specifying
filenames more efficient. So Linux commands can make use of special data streams to move input and
output to and from the command. STDIN can be though of as a gateway into
ls *.txt the code, STDOUT is a gateway for output. Note that there is a third
standard data stream called STDERR (standard error) which commands
will match all files ending in .txt. So important.txt, 123.txt, .txt often use to print error messages and warnings. We won't mention STDERR
would all be matched but data.txt.1 would not. again today. By default STDOUT gets routed to the screen display. It is also
easy to connect a commands STDIN to the keyboard device for example.
ls ?.txt The STDIN allows a program to ask you questions and you can type
responses.
will match all files with a single character before the .txt. So this
would match 1.txt, A.txt, +.txt but not 12.txt and so on.

ls [a-z].txt

will match all files of the form a.txt, b.txt, ... z.txt. Note, too that
wildcards can be used together so

ls [a-z]*.txt

will match all files beginning with a-z and ending with .txt and
containing any characters in between.
 Redirection Pipes

Linux can redirect standard output to targets other If commands use STDIN and STDOUT, is it possible to
than the display such as files by using the > character. connect the STDOUT from one command to the STDIN
of another?
YES – use pipes: |
FILE stdout stdout
Command Command Command Command
stdin stdin
Or take input from sources other than the keyboard
by using the < character. A simple example
du -sk /usr/bin/* | sort -n | tail -5
FILE Command
displays the five largest files in the /usr/bin directory

Redirection is a way of “grabbing” STDOUT or STDIN and forcing it to go Connecting commands together with pipes is one of the most powerful
somewhere other than the default. The most common instance is to features of Linux. Linux does not have a command to count the number
redirect STDOUT into a new file. This is extremely useful because it of files in a directory but it does have one command to list the files (ls)
means you can run a command and save the results automatically. To and a second command (wc) to count the number of lines in a list.
redirect STDOUT use the “>” symbol followed by the target so command
> file.txt redirects the output from command into a new file called file.txt. You could therefore use a pipe (|) to glue ls and wc together:
To redirect STDIN into a command use “<” like this: command < file.txt .
ls | wc -l

Note that we are using the -l (that is a hyphen followed by the lower
case letter l) argument or option for wc. To get information about any
command use

man command
eg
man wc
Exercises for Session 2

●
Using the command line to find your way
around. Session 3
Editors,
●
Using commands to manipulate files. regular expressions,
●
shell scripts
Getting help.
●
Using wildcards to match files and directories.
●
Searching and sorting.
●
Pipes and redirection.
●
Building your own commands.
 Controlling the command line
Text Editors
●
Use & to run commands in the background.
Many files in Linux are in plain text format: What happens if you forget?
●
Configuration files. ●
There are control-key commands which allow you to to
●
Shell programs. manipulate the command line:
●
Manual pages (although sometimes stored in CTRL-C interrupt a running program.
compressed format).
CTRL-D send an end of file, ending text input for most
So we need a utility which to let us change text files.
Linux/Unix programs.
●
emacs, xemacs
CTRL-Z suspend a running program.
●
vi, vim
●
gedit, pico, nano
Common to run an editor in the background [demo]

The choice of text editors provokes almost religious feelings among some These are very useful ways of breaking out of a command that looks
Linux users. There isn't a “right” editor to use. Generally a graphical editor like stuck. For a full list see
gedit or emacs is the friendliest to use. Editors like vi are considered much http://web.cecs.pdx.edu/~rootd/catdoc/guide/TheGuide_38.html
less friendly, but are much quicker for small quick changes to a file.
Note that CTRL-C can also be used for copying information. This can
Putting something into “background” means it continues to run, but is hidden cause confusion.
away behind the scenes and is not interactive like a foreground job. Adding an
ampersand after a command makes it run in the background. This is useful for
scripting because otherwise, you would have to close gedit every time you
wanted to stop and try your code out. By putting gedit in background, you can
still use your command line at the same time.
 Some editors gedit

vi – visual editor, now often known as vim (vi improved). GNOME Editor is widely available in Linux
●
Very fast, supports regular expressions and a
Simple text editor for the GNOME desktop.
relatively small but powerful command set.
●
● Intuitive – similar to notepad.
Almost all beginners find it difficult to learn.
●
GNU emacs - “extensible, customizable, self- Supports tabs so several files can be edited at
documenting real-time display editor”. once.
● ●
It is an “ integrated environment”: so can do much Context aware – highlights syntax.
else apart from editing.
But if you are familiar with another editor do use
●
It is aware of what sort of file you are editing, so that.
particularly useful for editing programs and scripts.

vi (pronounced vee-eye by purists) can be considered the first real screen gedit is a widely-available and simple-to-use text editor. It has fewer extra
based editor. Before that we used editors which allowed you to alter only a features than emacs and is easier to get started with than vi.
single line at a time. This was very difficult!
It is possible to add plug-ins to increase the flexibility and features available
vi is a “small” editor; that is it has relatively few commands and, in line in gedit. Use
with much Unix philosophy provides a single, specific function. There is no
spell checker or paragraph filler as these features are provided by other Edit -> Preferences -> Plugins
commands.
to find out more.
It is however not particularly intuitive to use and initially needs some effort
to understand.
emacs (properly GNU emacs) is a very powerful editor. What does
powerful mean here? The editor has many features and indeed is
sometimes referred to as an integrated environment rather than just an
editor. As well as editing files it is possible to read and send email and
compile and run programs from within emacs. It is also highly
customizable.

emacs is initially easier to learn than vi but its size and complexity can
mean it is easy to get overwhelmed by choice.
 Regular expressions Extended regular expressions

What is a regular expression? Extended regular expressions allow a much richer

● pattern to be specified.
Searches for string patterns within text files.
● Many powerful features
Uses metacharacters to extend a pattern.
●
Examples: Match white space [space, tab]
●
grep green fruit Match word boundaries
●
to search for green anywhere in the file, fruit. Digits only
●
grep ^green fruit Ranges

to search for green at the beginning of lines only. and many more.

These are both simple regular expression searches. When using regular expressions, extra options to grep are needed. For
example,
Regular expressions have some similarities with filename wildcards, but
are not the same. Regular expressions are used to find patterns within grep -E '\bgreen\b' fruit
text files, wildcards are used to match filenames.
would find all occurrences of green on its own, but not where it is joined to
another word. So

green cabbage

would match, but

greengage

would not.
 Some limitations
Taking control of the shell

Extended regular expressions can be used with many

So far we have looked at single commands:
commands.
●
For example we have run simple commands to do
However, the way these are formed varies with one thing, e.g. cd, mkdir, grep and so on.
different applications.
More typically:
The most common are variations are ●
Jobs are more complex than the ones we've seen.
● ●
Command line tools (grep, sed and so on) Need recipes of multiple commands.
● So we need to control the flow of :
Perl
●
● Commands.
C#
●
●
Data.
PCRE – Perl Compatible Regular Expressions

For our exercises today, we don't need to worry about these differences.
Many of the metacharacters and search terms are the same. However, if 'Real World” problems are often too complicated to solve with a single
you are using regular expressions in different applications it's worth command. They also need data to be read and written in a controlled
checking whether there are differences or limitations. way. This session explores some of the ways this can be achieved with
a Linux command line interface.
Note that C# is the Microsoft implementation of ; it provides very good
support for regular expressions.
 Scripts Writing scripts

● ●
Rather than typing commands on the command It's usual to write some of the script,
try it out, save a working version Use gedit
line it can be more convenient to put the to create
and add to it. This cycle is repeated. the first
commands into a single file. This is called a version of
the script.
●
Rather than closing your editor Save it.
script.
every time you want to try the script,
● run it in the background with “&”
A script can be run more simply than a series of
Run the
● script
commands which you have to get exactly right For example
every time. gedit &
emacs &
●
You keep control of the command Make
changes
line and fix
errors. Save
script.

In the exercises, we recommend you write a script and try it out until it
Scripts are just plain text files containing commands and optionally, but works. When you're happy with it, save it under a different name (with a
usefully, flow control. Flow control is the means of determining the order in version number on the end for example). This way, if you make a mistake
which commands are executed. in the next exercise, you can go back to the earlier saved version and start
again.
 When to script When not to script

● ●
A lot of what you want to do is covered by Your problem can't be broken down into
one or more Linux commands anyway. existing UNIX commands.
● ●
The problem is conceptually simple. You need lots of data/program flow
●
control.
It doesn't have to be blindingly fast.
●
●
It has to be fast.
It's something you have to do a lot.
●
●
It's very numerical.
It's something you want to be automated.
●
It's a one off job.

Once you are familiar with the technique, shell scripts are quick to write. We have already seen that shell scripts can be seen as a quick and dirty
They are often used to avoid tedious and repetitive tasks – for example in solution. With this in mind any problem that requires a carefully designed
the exercises we will write a short script to create several files with similar solution is not a good candidate for a shell script.
names. They are used, too, automate various system administration tasks
that are too trivial to merit a program but for which there is no existing Any application that needs to be fast or involves number crunching or large
command. scale data manipulation should not be solved by a shell script. A program is
needed.
Shell scripts don't require the investment of time and effort that programming
in a higher level language such as C or C++ or Perl or Python. Scripts are Any application that needs a structured solution involving subroutines and
often characterised as “quick and dirty” but there is, I think, a place for this functions is too big a problem for a shell script.
sort of solution.
Executables Exercises for Session 3

●
In Windows .exe files are executable – in ●
Regular expressions
Linux the conventions are much looser. ●
Use the gedit text editor
● ●
In Linux if you write a script you have to Write some shell scripts
manually give people permission to run it.
●
The chmod command changes
permissions on a file:
chmod +x file
allows your script to be executed.
 Session 4

●
Using computers remotely.
●
Session 4 We will also look briefly at package
Using computers management and system administration.
remotely
●
Exercises.

The computer we want to use is not always the one in front of us. How
does Linux manage this?

We will also look at the privileged account – root – which is there to

enable us to make changes to the system and how systems are managed.
 Using remote systems Accessing systems and copying data

● ●
Often the computer we are sitting at is not the Use
only one we need to use.
ssh
– More powerful
to access remote computers.
– Shared resources ●
Use
– Additional software
scp
●
Set up your Oxford Single Sign-On (SSO) to
access the IT services linux system. to move data.

For users of ARC – the academic research computing facility – and for The ssh command gives us a secure, command line session on a remote
many others, the computer in front of you is not the most powerful or machine.
useful computer available.
In some cases,
We need to find ways to access these remote computers.
ssh -X remote-system
IT services provide a linux commmand line service. The system is running
(at time of writing) Debian 7.11, or the final version of Debian Wheezy. will allow you to open remote graphical applications on your local system.
Although this is a different Linux distribution from the one we’re using, the So for example,
command line interface is generally consistent between distributions.
ssh -X [email protected]
All members of the University can use this system.
then

gedit &
 Managing jobs on remote systems

●
Why?
●
Jobs may last longer than a few minutes.
● Managing systems
Don’t want to keep a session on a remote
system: lose connection, lose job.
●
Use
screen
to detach from a session and logout.

There are alternatives to screen, for example tmux. Both commands

perform the same function – they allow you to leave a session running on
a remote system while disconnecting your local session.
Managing systems Why manage packages?

The following slides discuss briefly techniques ●

Complexity of current systems.
for ●
Dependencies – new applications may affect existing
●
Managing Ubuntu systems software.
●
– Using a privileged account Each system may be configured slightly differently to
suit its own environment.
– Maintaining a system: updates
●
Enormous range of hardware that must be supported.
●
Installing new software: different methods ●
We don't want – if possible – to concern ourselves with
– Using the Software Center these details when we need to install some new
– packages.
Using the command line
–
Additionally, as with Windows a mechanism is needed to
The difficult way! keep the system up-to-date.

Why do we need package managers.? We have already seen the enormous

range of software available in Linux.

Developers cannot predict always what other packages will be used on the
same system as their software; how the system is set up or what version or
distribution is being used.

Some method of managing these problems was needed. Both package

managers keep a database of information about software installed on the
system which is used as a reference when a new package is installed.

A means of keeping a system up-to-date is also needed. You could view these
package managers as similar to Windows Install Shield.
 Ubuntu Software Updater Other ways of installing software

● The quick way:

Provides graphical access to un-installed but
available packages. ●
Using the command line
●
Can also be used to update the system. apt-get install package
●
Unfortunately not working correctly on this Live The hard way:
Ubuntu release. ●
Software is not always bundled in a format that
● APT or RPM can use.
Will work fine on installed version.
●
It will come in a single .tgz or .tar.gz or .bz2 file.
●
Also known as a tarball.
●
Requires rather more knowledge of the system.

APT is a rather more flexible and powerful version of RPM and yum. It can Until now we have looked at software that has been easy to download and
also deal with software in different formats. install. We have not had to concern ourselves with the details of what
version we need or where to go to download the software.
I have little experience of setting up an APT system from scratch but this
version of Ubuntu comes configured to run apt successfully. But there will be times when this is not possible so that we will have to find
the software ourselves. It will then need configuring, building and installing
by hand.

One of the many disadvantages with this system is that it is then much
harder to remove the software afterwards as it is not always obvious
whereabouts on the system all the files have been copied.
 Managing your system The root account

● ●
Until now we have used the the system to In most Linux systems you will have your own
change our files but not made any system username and password.
changes. ●
Standard user accounts are set up so that you
●
Unlike Windows, Linux has a clear and rigid can change your own home files but not any
distinction between using a system and making system files.
changes to a system, such as installing new ●
To install new software we need an account that
software.
can change files anywhere on the system.
●
So to make changes a special account is ●
This account is known as the root account.
needed.

This “separation of powers” is fundamental to the way Linux (and Unix) Note that because Ubuntu is running from USB separate user accounts are
systems are designed. As you may remember from an earlier session not configured.
you were able to move around the whole system but could not remove
files apart from those in your home directory. To run commands as root you need to use

sudo command

in a command tool Window and press enter. You should now see a different
prompt: a # . This a standard Unix/Linux convention for indicating that this
account is privileged and so dangerous.
 Using sudo
Exercises

●
Some systems use a different approach to
We will use the IT services linux service to:
administering system.
●
Use sudo + command when carrying out ●
Log in remotely.
privileged actions. ●
Copy files between systems.
sudo apt-get install bluefish ●
Demonstrate a technique to manage jobs on
will install the bluefish editor. other computers.

●
Normally you will be prompted for a password;
this will be remembered by the system for about
15 minutes.

Ubuntu developers decided that this mechanism is more secure than a

separate root account.

There are advantages and disadvantages to this approach. It is probably more

secure: users have one less password to remember. A log is kept of all sudo
activity – you can check it later when you are doing the exercises – have a
look in /var/log/auth.log. It is also possible to configure sudo so that a
user can only do a limited range of privileged commands.

A disadvantage of this approach is that it doesn't work particularly well for

desktop systems in a network where most accounts are managed by a central
server. However, this shouldn't concern home users.
76