Study Notes GNU Linux Command Line Interface

STUDY NOTES
GNU/Linux: the command line interface

(A GPL’d Document)
Anthony A. Aaby
DRAFT Version 0.9. Edited February 25, 2004

This material is copyrighted under the following copyrights:
Copyright c 2003 by Anthony A. Aaby
Walla Walla College
204 S. College Ave.
College Place, WA 99324
E-mail: aabyan@wwc.edu
LaTeX sources available at http://cs.wwc.edu/~aabyan/Linux/linux.tgz
This is a free book; you may mirrow it, reproduce it, and/or modify it under the terms of version 2 of
the GNU General Public License as published by the Free Software Foundation whose website is located
at http://www.gnu.org/copyleft/gpl.html.
This book is an edited and expanded version of Linux Fundamentals: A Training Manual by Philip
Carinhas, Ph.D. of Fortuitous Technologies (www.fortuitous.com, pac@fortuitous.com). The original
copyright is reproduced here:
Copyright c 2000-2001 Fortuitous Technologies, Inc.
Fortuitous Technologies Inc.
6909A Hardy Dr.
Austin, Tx 78757
USA
WWW: http://fortuitous.com
E-mail: fundamentals@fortuitous.com
This training manual is a free book; you may reproduce and/or modify it under the terms of version 2 of
the GNU General Public License as published by the Free Software Foundation whose website is located
at http://www.gnu.org/copyleft/gpl.html.
This book is distributed in the hope it will be useful, but without any warranty; without even the implied
warranty of merchantability or fitness for a particular purpose.
The author encourages wide distribution of this book for personal and commercial use, provided the
above copyright notice remains intact and the method adheres to the provisions of the GNU General
Public License located at http://www.gnu.org/copyleft/gpl.html.
In summary, you may copy and distribute this book free of charge or for a profit. No explicit permission
is required from the author for reproduction of this book in any medium, physical or electronic.
Note, derivative works and translations of this document must be placed under the GNU General Public
License, and the original copyright notice must remain intact. If you have contributed new material to
this book, you must make the source code (e.g., LATEX source) available for your revisions.
Contents
1 Introduction to GNU/Linux 5
1.1 Linux Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2 Multi-User Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.3 Why GNU/Linux? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.4 User Interface: Graphical User Interface vs Command Line Interface . . . . . . . . . . . . 7
1.5 Conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2 Using X-Windows 9
2.1 X Display Manager, Window Managers and Desktops . . . . . . . . . . . . . . . . . . . . 9
2.2 Starting and Customizing the X Environment . . . . . . . . . . . . . . . . . . . . . . . . . 10
3 UNIX Command Line Basics 15

3.1 Logging In To Your Account . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.2 Command Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.3 The Linux Manuals and the man Utility . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.4 File System: Create, List, Copy, and Move . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.5 I/O, Redirection, and Pipes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.6 Command Line Editing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4 Text Files 27
4.1 Command Line Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
4.2 Vi and related editors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
4.3 Editing With Pico . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
4.4 Editing With emacs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
4.5 Using Mail . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
4.6 Regular Expressions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
5 The Linux Environment 39

5.1 The UNIX Shell Game . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
5.2 Bash . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
5.3 Shell Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
5.4 Choosing the Right Path . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
5.5 Groups and Newgrp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
6 User Level Security 47

6.1 Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
6.2 Social Engineering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
6.3 Cryptography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
6.4 Authentication and Integrity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
6.5 Privacy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
DRAFT COPY February 25, 2004 3

7 Filesystem Essentials 49
7.1 The Linux Virtual Filesystem (VFS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
7.2 File Attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
7.3 Changing File Attributes with chmod . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
7.4 Changing File Ownership with chown and chgrp . . . . . . . . . . . . . . . . . . . . . . 54
7.5 Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
8 Managing Filesystems 57
8.1 Creating Partitions with Fdisk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
8.2 Creating a Filesystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
8.3 Mounting Filesystems by Hand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
8.4 Swap Partitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
8.5 Automatic Mounting at Boot with /etc/fstab . . . . . . . . . . . . . . . . . . . . . . . . . 60
9 Process Control 61
9.1 Creating Foreground and Background Processes . . . . . . . . . . . . . . . . . . . . . . . . 62
9.2 Killing Processes With kill . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
9.3 Managing Process Priorities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
9.4 Cron . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
10 Shell Scripting 69
10.1 Shell Initialization Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
10.2 Utility Shell Scripts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
10.3 General Scripting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
10.4 Script Automation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
11 Networking and The Internet 75

11.1 Network Communication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
11.2 Using X-Windows Over the Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
11.3 Network Security for Users . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
12 Program Development in Unix/Linux 83

12.1 C & C++ Compilers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
12.2 Debugging with gdb . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
12.3 The Unix make Facility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
12.4 Concurrent Version Control (CVS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
12.5 Packaging Your Software - Tar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
12.6 Session Capture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
12.7 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
A Glossary 93
B Unix/Linux Commands by Location 95
C Unix/Linux Commands by Category 101

Chapter 1
Introduction to GNU/Linux
Linux is a computer operating system kernel originally developed by Linus Torvalds as a

research project. Its rapid evolution to a full featured operating system in less than a decade is
remarkable. Linux runs on a variety of hardware platforms including those from Intel, Apple,
Sun, and others. The Linux kernel packaged with GNU software from the Free Software
Foundation is available in a variety of distributions. The best known include Red Hat, Suse,
Mandrake, Gentoo, Debian, and Slackware. See www.linux.org and www.distrowatch.com
for more information on various distributions.
Linux is based on UnixT M which was developed for programmers by programmers. This work
is designed to introduce computer science, computer engineering, and software engineering
students to Linux command line interface.
Only selected commands and features of the command line interface are covered. The user
is assumed to need little more than an overview and a collection of useful examples. The
material presented is more than sufficient to enable the reader to effectively and efficiently
interact with a Unix/Linux environment
1.1 Linux Features

The GNU/Linux distributions provide an operating system that
• is a full-featured, 32-bit multi-user/multi-tasking OS.
• adheres to the common (POSIX) standards for UNIX .

• provides native TCP/IP support.
• has a mature X Windows GUI interface.
• provides a complete development environment with C, C++, Java, editors, version control systems,
and built tools.
• is fully Open Source.

CHAPTER 1. INTRODUCTION TO GNU/LINUX
1.2 Multi-User Operation

In UNIX and Linux all interactions with the OS are done through designated “users”, who each have an
identification ID (login name) and a password. UNIX allows different users to co-exist simultaneously and
allows for different levels of users.
The most powerful user is called superuser or “root”, and has access to all files and processes. The
superuser does many of the system management tasks like adding regular users, file backups, system
configuration etc.
Common users accounts, which perform non-system type tasks, have restricted access to system-sensitive
components to protect Linux from being accidentally or purposely damaged.
1.3 Why GNU/Linux?

Linux can operate as a web, file, smb (WinNT), Novell, printer, ftp, mail, SQL, masquerading, firewall,
and POP/IMAP server to name but a few.
It can act as a graphics and multimedia (audio, video), development (C, C++, Java, Perl, Python, SQL),
and office workstation etc.
Linux
Workstation Server
Firewall
Documentation SQL
Communications
Video Samba
File
Audio
C/C++/Java/Perl/DB FTP Proxy
Graphics Httpd
Figure 1.1: Linux Uses
Linux is a good solution for developers that need a stable and reliable platform that has open source code.
Linux has many GUI software development interfaces. Linux is ideal as a workstation also, and offers
many customizable features not found in any other platform. It makes a good platform for dedicated
workstaions that have limited functions like in an educational or laboratory environment.
6 DRAFT COPY February 25, 2004

1.4. USER INTERFACE: GRAPHICAL USER INTERFACE VS COMMAND LINE INTERFACE
1.4 User Interface: Graphical User Interface vs Command Line

Interface
Unix supports both command line and graphical user interfaces. In some versions of GNU-Linux, the
key combination, Ctrl Alt F6, switches the X based GUI mode to terminal mode with its command line
interface. In terminal mode, Ctrl Alt F2 switches the display to the X based GUI mode. A terminal
window, launched from within the X display, may be used to learn the command-line interface.
1.5 Conventions
In order to take full advantage of this manual, students should execute every command that is listed in
the text as well as do all the exercises. The following is a list of conventions supported in this manual:
<bash>: indicates a command entered in the bash shell by the user. When you see this sign, you are
expected to enter these commands exactly as indicated and check that the results are consistent
with what is written.
<tcsh>: indicates commands in the tcsh shell. Most of these commands are to be completed after hours or
at home. Since Linux advocates freedom of choice, we wish to make students aware of this option
to bash.
<Super>: indicates a command entered by the System Administrator or Root.

Bold indicates a command entered at the prompts above.
Big Bold TW is used to identify commands in the text.
Underline indicates a file or a directory in the text.
Slant indicates command options.
Plain TW indicates a screen text of command output or editor.

CHAPTER 1. INTRODUCTION TO GNU/LINUX

Chapter 2
Using X-Windows
The X-Windows system is the Linux answer to a graphical user interface or GUI. It allows
users a great interface to work from.
The X Window System (www.x.org), known simply as “X”, is a portable, network-transparent
client/server interface between hardware and the desktop environment. It is frequently used
in conjunction with the UNIX and Linux operating system. The current version of the X
Window System is “X11R6”. A freely available version of the X protocol for the Intel platform
is XFree86 (www.xfree86.org).
If your system is running X-Windows, there should be a prompt to enter a user name and
password. As the focus of this text is on the command line interface, this chapter is here for
completeness and may be skipped.
2.1 X Display Manager, Window Managers and Desktops

An X display manager is used to provide login services and session management. The X environment
has many desktops and window managers that completely control the look and feel of the environment.
A window manager is the component which controls the appearance of windows and provides the means
by which the user can interact with them. Virtually everything which appears on the screen in X is
in a window, and a window manager quite simply manages them. A desktop environment aims to
provide a more complete interface to the operating system, and supplies a range of integrated utilities
and applications. Two freely available desktops are KDE (K Desktop Environment, www.kde.org) and
GNOME (www.gnome.org). KDE is supported by the Kdev window manager while Gnome is supported
by a variety of window managers such as MetCity, Sawfish, Englightment, and IceWM.
Here is a list of the most popular desktops, window managers, and display managers:
Desktops
CDE Uses the Motif libraries.
GNOME Uses the GTK+ libraries.
.gnome Configuration files.
KDE Uses the QT X-library.
.kde* Configuration files.
X-Window Managers for Linux

CHAPTER 2. USING X-WINDOWS
AfterStep NextStep feel and beyond

Enlightenment An extreme, detailed, and configurable environment.
FVWM Young nephew of TWM
.fvwmrc Configuration files.
FVWM2 Clever brother of FVWM
.fvwm2rc Configuration files.
IceWM A fast and flexible window manager.
Kwin Window manager distributed with KDE.
MWM Motif window manager.
MetaCity Window manager distributed with Gnome. Built using
Glade and the GTK+ tool kit.
OLWM Open Look window manager. Once the standard win-
dow manager from SUN.
Sawfish An extensible window manager.
TWM One of the oldest and simplest managers. Very basic.
.twmrc Configuration files.
Window Maker NextStep look and feel. Easy to configure
X-Display Managers
kdm KDE display manager
gdm Gnome displays manager
xdm X display manager
The following subsections provide a brief description of the desktops.
2.1.1 Gnome
Gnome is one of the two very popular desktops out there.
Lets look at Figure 2.1. First notice that the bottom panel has several typical icons like a tool-box,
terminal window, and netscape.
It has a four panel virtual display which gives you four whole screens to work in. This gives some room
to spread out. Inside the window we see some drop down panels that set the appearance. There is also
the file manager visible. There are click-able icons on the desktop itself, much like the other OS’s you
have used. Not too shabby for a free OS.
2.1.2 KDE
KDE is the other popular desktop. Figure 2.2 is a sample of what the KDE desktop looks like.
Notice that KDE also has four virtual panels to work on, and has all the bells and whistles a high quality
GUI would provide.
2.2 Starting and Customizing the X Environment

If X-Windows is installed and no X-display manager is running on your system, start X from the console,
by typing
startx
If everything works, you should have a terminal that has a few big black windows and a clock or two.
This is just a single version of how X can be presented. X has many toys to play with. Here are a few:

2.2. STARTING AND CUSTOMIZING THE X ENVIRONMENT
Figure 2.1: Gnome Interface
Basic X Tools
xterm [options] Start an X-Terminal
xclock [options] A basic wall clock
xman -options Display the manual X-style
xbiff -option Mailbox flag to keep you waiting for mail
xboard A chess game
xrdb -merge Xfile Incorporate Xfile into your environment
xload Show the system’s usage load
There are actually too many applications in X to list. Almost everything these days has an X interface.
Just about everything in X can be customized, depending on how hard you want to work on it. The
most simple thing you can do to customize your environment is to keep a defaults file, normally called
.Xdefaults located in your home directory.
A typical .Xdefaults looks like this:

Figure 2.2: KDE Interface
!---------------------------------
! .Xdefaults X-Resource File
!---------------------------------
Netscape.Navigator.geometry: =700x700
XTerm*visualBell : on
XTerm*Font : 9x15bold
XTerm*scrollBar : on
XTerm*saveLines : 4940
XTerm*borderColor : white
xclock*geometry : 60x60-0+0

2.2. STARTING AND CUSTOMIZING THE X ENVIRONMENT
xclock*background : black
xclock*foreground : white
The initial instance in the variables is a class name. The subsequent variables in string are called X-
resources or just Resources which identify the property of the program or property to modify. The *’s
are wildcards and can match anything.
Finding what is and isn’t a valid X-resource can be difficult, but normally most man pages have a listing
of these and examples of how to set them. Once you have edited this file to your liking, you must activate
the changes by using the xrdb utility like this:
<bash>: xrdb -merge .Xdefaults
Afterwards, try a command to see if your changes take hold. Normally, there is an initialization script in
a system folder or in your home directory that will read your .Xdefaults for you when you start your X
session. This file is sometimes called .xinitrc and is easy to customize:
#!/bin/sh
userresources=$home/.Xdefaults
usermodmap=$home/.Xmodmap
sysresources=/usr/X11R6/lib/X11/xinit/.Xresources
# merge in Xdefaults and keymaps

if [ -f $home/.netscape/lock ]; then
rm -rf $home/.netscape/lock
fi
if [ -f $sysresources ]; then
/usr/bin/X11/xrdb -merge $sysresources
fi
if [ -f $sysmodmap ]; then
xmodmap $sysmodmap
fi
# start some nice programs

xterm &
xclock &
# fix the bell- 11 Nov 96 rcr
/usr/bin/X11/xset b 50 4500 100 &
exec fvwm


Chapter 3
UNIX Command Line Basics
This chapter introduces the commandline interface and commonly used commands.
3.1 Logging In To Your Account

Access to your account may be through a terminal window or a X based login screen. The process of
logging into the system involves the use of a user name and a password.
Text-based interfaces occur in five different contexts. They are:
1. X-Windows GUI interface. Open a shell or terminal window to use the command line interface.
2. Text-base terminal: A single use (at a time) Unix/Linux system or a text-based terminal on a local
area network (LAN).
3. Dialup/PPP connection: A connection to a remote computer through a dial-up interface. The local
machine provides either telnet or ssh to provide login to a text window interface.
4. Local area network (LAN) connection: The local machine provides either telnet or ssh to provide
login to a text window interface.
5. Internet connection: A local host provides access to a remote host through telnet or ssh.
<login:>: YourUserName
Password: - for security, nothing will be displayed.
When you log into your system with the login name and password given to you, you will see a prompt,
something like
<bash>:
You may see a variety of prompts depending the shell your are using and the set prompt command in
the .login file for the c-shell. Your system should always prompt you with the name of the shell (bash)
and your login name. This is a customizable feature in the bash shell which you are now using.
To logout enter the command logout, exit, or a period.

CHAPTER 3. UNIX COMMAND LINE BASICS
<bash>: logout
or
<bash>: exit
or
<bash>: .
You may change your password with the passwd command
The most basic command in Linux is the directory listing “ls” command. You can see the contents of
your account by typing
<bash>: ls -al
You should see files like
drwxr-xr-x 13 joe user 1024 Nov 9 17:06 .dt

-rwxr-xr-x 1 joe user 5111 Sep 30 15:19 .dtprofile
drwxr--r-- 2 joe user 96 Dec 1 12:25 .elm
-rwxr--r-- 1 joe user 1178 Nov 19 10:58 .emacs
User level filesystem commands are covered in Section 3.4 and the essentials of the filesystem are covered
in Chapter 7.
3.2 Command Structure

We’ve already seen ls which gives a directory listing of the current working directory (cwd). Commands
in Linux follow a general format:
Command -options Other parameters
for example:
<bash>: ls -l .bashrc
First comes the command name, followed by options. Options are normally preceded by a dash or minus
sign. There is always a space between the command and the dash. Some commands use no options at
all. After the options comes any other parameters or informations that command may need.
Let’s talk about some of the workhorse commands. Please note that these definitions are purposely
abbreviated and incomplete! For more details on each of the commands below, you can consult the online
manual using the man and info commands, see Subsection 3.3. In the commands below, parameters
that are enclosed in square brackets [...] are optional to that command. [-opts] refers to options in the
style just mentioned.
Special Keys Strokes:

q quits from many commands like more, less, and man
Ctrl+c also quits out of many commands
Ctrl+l clears the screen
Ctrl+a puts the cursor at the beginning of the command line
Ctrl+e puts the cursor at the end of the command line
Ctrl+z suspend the current process

3.2. COMMAND STRUCTURE
Get Help and Information

man cmd Consult the manual page for cmd
man -k keyword List man pages with keyword (same as apropos key-
word)
info keyword List info help pages containing keyword
grep pattern dir look for pattern in the file in dir
grep pattern file look for pattern in file
env [-opts] [exp] Print environment or run a command with another envi-
ronment.
find [path] [exp] Find files in path using exp
locate keyword Locate all files of name keyword in a database
printenv Print environment variables (see set)
set [vars] Print/Set environment vars and functions
whatis keyword Search the whatis DB for keyword
whereis command Locate source/binary and manuals for command
which command Display path of command
Text Files and Printing

cat file(s) Concatinate files, send result to standard output.
cat file Display contents of file without paging
more file Display & page the text file (See less)
less file Display & page the text file
head file List the first few lines of file
tail [-opts] file List the trailing lines of file
diff A B List the differences between A and B
grep pattern file Find lines containing pattern in file
wc files List the number of characters, words, and lines in files
clear Clears the screen. Same as Ctrl+L
tr chars1 chars2 file Change chars chars1 in file to chars2
awk| gawk [pgrm][file] Filter file by pgrm
sed [script] file Stream edit/filter file using script
Printing
lp file System V print command
lpr [-Pprinter] file Print file using printer. See /etc/printcap for list of print-
ers.
lp -Pprinter file Print file using printer
lpr -Pprinter file Print file using printer
lpq [-Pprinter] Check the print queue
lpstat System V: Check print queue
lprm [-Pprinter] job# Remove job from printer queue
cancel job System V: Remove job from print queue
General Tools
. Current working directory
.. Parent directory
User’s home directory
cd dir Change cwd to dir (home if dir omitted)
cd .. Change cwd to parent directory
cd /path Change cwd using absolute path
cd path Change cwd using relative path
chmod perms files Change file permissions of files

chown owner.group files Change file owner and/or group

chsh Change the default shell
cp [-opts] f1 (f2—dir) Copy file f1 to f2 or directory dir
date Displays the date
du -h Report disk space used by the current directory and sub-
directories
kill pid Kills process ID pid
ln [-opts] Old New Link Old to New
login [username] Login to system with UID username
lpr file Print file on the default printer
ls [dir] Listing for dir (cwd if directory is omitted)
ls [-l] List contents in long (detailed) format
ls [-a] List hidden files as well
mkdir dir Creates directory dir
rmdir dir Remove directory dir. dir must be empty
mv file1 file2 Rename file1 to file2
passwd [-opt] username Change password
ps [-opts] Output a list of currently active processes
pwd List the path to the current working directory
rm files Removes files
rm -r * Recursively remove all files and directories in cwd
startx Start the X-Windowing system
tar [-opt][arch][file] Manage tar archives
telnet [host [port]] Connect to the remote host
uname [-opts] Output name and version number of OS
who List users logged into this system
xterm [-opts] Start a brand new X-terminal window
3.3 The Linux Manuals and the man Utility

Virtually every command that is worth knowing has an entry in the man pages, and is accessed by doing
a man command . To get all related commands to a keyword word, use man -k keyword as in
the following example:
<bash>: man -k manual

man (1) - format and display the on-line manual pages
man2html (1) - format a manual page in html
perlxs (1) - XS language reference manual
whereis (1) - locate the binary, source, and manual page files
xman (1x) - Manual page display program for the X Window System
When in doubt, use man and man -k keyword to get info on a command or UNIX related term.. Find
the man page for the ls command. ls will list directory contents:
<bash>: man ls
Here is a man entry for cd which changes your current working directory (folder).
<bash>: man cd

3.3. THE LINUX MANUALS AND THE MAN UTILITY
LS(1) FSF LS(1)

NAME
ls - list directory contents
SYNOPSIS
ls [OPTION]... [FILE]...
DESCRIPTION
List information about the FILEs (the current directory by
default). Sort entries alphabetically if none of -cftuSUX
nor --sort.
-a, --all
do not hide entries starting with .
...... etc ........
Figure 3.1: Manual Page for ls
cd(n) Tcl Built-In Commands cd(n)

NAME
cd - Change working directory
SYNOPSIS
cd dirName
DESCRIPTION
Change the current working directory to dirName, or to the
home directory (as specified in the HOME environment variable)
if dirName is not given. Returns an empty string.
Figure 3.2: Manual Page for cd
3.3.1 Exercises
1. <bash>: man man
2. <bash>: man 7 signal
3. Man cd and look for information on “.” and “..”
4. <bash>: cd .. ; pwd
5. <bash>: cd . ; pwd
6. <bash>: cd ; pwd
7. Discusss what “.” and “..” are in terms of the filesystem.

3.4 File System: Create, List, Copy, and Move

The fundamental file system commands for files and directories are those to list, create, copy, move
(rename), and remove files. This section will show you the basics. Lets start with creating some files
now, so we can list them later.
3.4.1 Creating New Files

File creation can be done by invoking an editor on a new filename. Before we try this, there is an easier
way to make a new empty file by using the touch command. This is how it works:
<bash>: touch file1
This will make a new file named file1 that is empty. There are many other ways to make new files that
we will see later.
3.4.2 Creating New Directories

Directories are created with the mkdir command. Thus
<bash>: mkdir dir1
will create a new directory named dir1 located in your current working directory. List this file now with
the “long” format we just spoke about. Check your files with
<bash>: ls -l
3.4.3 Listing Files and Directories

By far, listing a file is the most basic of all commands. As we have seen before, you list files with the ls
command:
<bash>: ls file1
file1
<bash>: ls -l file1
-rw-r--r-- 1 joe users 0 Oct 28 22:05 file1
Using the -l flag causes a “long” listing that shows more information about file1.
You list your directories in a similar way (note that the second example shows that dir1 is empty).:
<bash>: ls -l
total 10
drw-r--r-- 1 joe users 0 Oct 28 22:05 dir1
<bash>: ls -l dir1
total 0
3.4.4 Copying Files and Directories

You can copy a file by using the cp command. The following statement
<bash>: cp file1 file2

3.4. FILE SYSTEM: CREATE, LIST, COPY, AND MOVE
will copy file1 to file2. Copying a folder or directory requires the use of the recursive or -r flag
indicating that cp should decend into the directory and copy all sub-files and sub-folders:
<bash>: cp -r dir1 dir2
Exercise 3.4.4: Please create a directory like this now. Explain clearly how the following examples
are different from the above:
<bash>: cp file1 dir1
<bash>: cp -r dir1 dir2/
<bash>: cp -r dir1 dir2/dir3
3.4.5 Moving Files and Directories

Moving a file is the same as renaming it. This allows for the possiblity that you move the file to another
directory as well:
<bash>: mv file2 file3

<bash>: mv file1 dir1/file2
You move a directory in EXACTLY the same way as a file:

<bash>: mv dir1 dir3
3.4.6 Changing Directories

Changing your current directory is done with the cd command:
<bash>: cd dir1
<bash>: cd ../dir2
The cd command without any options or parameters changes working directory to the users home
directory.
3.4.7 Path to the Working Directory

It is easy to get lost in a filesystem. The command pwd displays the path from the root directory to the
current working directory.
<bash>: pwd
3.4.8 Disk Space Usage

Disk space utilization is reported by the command du -h which reports the the disk space used in the
current directory and all subdirectories. The option -h appends a size letter to each size reported.
3.4.9 Removing Files and Directories

Normal files are removed with the rm command:
<bash>: rm file1

Removing directories is also done with the rm command, but if the directory is not empty, you need to
use the “recursive” or -r option:
<bash>: rm -r dir1
After you try the above, make sure file1 and dir1 are gone. Empty directories may be removed with
the rmdir dir1.
3.5 I/O, Redirection, and Pipes

I/O refers to Input (I) and Output (O). This section talks about input and output of commands and
how you can manipulate these. These prinicipals are very important because they are used constantly in
Unix.
3.5.1 Standard I/O

In UNIX , Standard Input (stdin) and Standard Output (stdout) are mechanisms that allow you to input
or output data from a command line. Simple commands like ”cat file1” send their results to stdout
(normally to your terminal screen) while the word file1 is an example of stdin which is fed to the
command cat.
Independent of stdin and stdout, there is the also standard error (stderr) which normally goes to your
screen when the command detects an error. Its manipulation is shell specific.
Bash assigns special numbers, called File Descriptors, to stdin, stdout, and stderr:
Name Abbreviation File Descriptor Standard Device

Standard Input stdin 0 Keyboard
Standard Output stdout 1 Console
Standard Error stderr 2 Console
Table 3.1: Bash Standard I/O
3.5.2 Redirection
Redirection refers to the art of redirecting input and output traffic from commands. Shells like bash
allow for redirection of stdin and stdout with the < and > operators respectively.
As an example, lets say you want to list some files and send (redirect) the output to a file instead of the
screen. Do it the easy way:
<bash>: ls -al > output.txt

To check the output, you can use cat (short for concatenate). cat is uselful when you want to view
short files:
<bash>: cat output.txt

-rwx------ 1 carinhas users 1606 Aug 17 23:14 .acrorc
-rwx------ 1 carinhas users 153 Dec 20 08:47 .bashrc
-rwx------ 1 carinhas users 3189 Dec 23 15:34 .cshrc
................

3.5. I/O, REDIRECTION, AND PIPES
Examples of stdin redirecting:
<bash>: cat < output.txt

<bash>: wc -l output.txt
7 output.txt
<bash>: wc -l < output.txt
7
Note that “cat output.txt” and “cat < output.txt” give the same result, but the ”wc -l”
examples give something slightly different.
In bash stderr is redirected with with the 2> operator, while in tcsh, the >& operator. Just relax and
we will see real examples of this shortly.
Linux Warning: Please note that > will overwrite anything in the output file, if it exists, or create the
file if it does not exist. In contrast, the >> operator will append to the existing file.
Here is a brief summary of the redirects:
Name Operator Description

Redirect stdin < Feeds the file to input
Redirect stdout > Creates or overwrites
Append stdout >> Creates or appends
Redirect stderr >& Both stdout and stderr
Redirect stderr 2> Only stderr in bash
Table 3.2: Standard Redirection I/O
3.5.3 Pipes
When you want to take the output of one command and use that is input into another, use the “pipe
operator” | . Think of actually connecting a metal pipe from one command to another. The following
example sorts a simple ls command in reverse order (do it!):
<bash>: ls | sort -r
.doomrc
.cshrc
.bashrc
.acrorc

3.5.4 Examples to Try
<bash>: cat noname 2> error.txt # Sends error to error.txt

<bash>: cat noname > error.txt 2>&1 # Send stdout + stderr to error.txt
<bash>: cat noname >& error.txt # everything (as above) goes to error.txt
Try these examples in tcsh:
<tcsh>: tcsh
<tcsh>: (cat noname > output.txt) >& error.txt # Send stderr to error.txt
<tcsh>: more output.txt # Same as in bash.
<tcsh>: more error.txt # Just like in bash.
<tcsh>: cat testfile.txt >& out.txt # ditto
<tcsh>: cat < testfile.txt | sort | more # ditto
<tcsh>: locate .bashrc | xargs grep alias # find ‘alias’ in .bashrc.
<tcsh>: exit
Don’t forget the last “exit” to get out of tcsh and back into fabulous bash.
3.6 Command Line Editing

A great shortcut in bash and tcsh is to use the command editing facilities. Just hit the up-arrow key
to a previous command and move the cursor to edit that command. Just try it. Command editing is
very useful in repeating and correcting previous commands.
Now we discuss other features of command line editing.
3.6.1 Command and File Completion

Most shells support command and file completion typing shortcuts. These shortcuts allow you to hit the
Tab key to finish off the name of a command or file after only hitting a few keys. Try this example of a
filename (directory):
<bash>: cd /usr/inc<TAB> will produce
<bash>: cd /usr/include/
Now try this example for a command:
<bash>: ghostv<TAB> should produce
<bash>: ghostview
Note that command and file completion can only complete upto a unique set of commands. This means
you have to provide a unique start string.
Thus <bash>: ghost<TAB> won’t work because it could complete to ghostview or ghostscript.
tcsh has the same file-name completion mechanism as bash. The <TAB> key will complete the filename
and command up to unique names.

3.6. COMMAND LINE EDITING
3.6.2 Possible Command Completion

bash will show the possible choices are by hitting <TAB>twice (<Ctrl-D> in tcsh ):
<bash>: ls /etc/h<TAB><TAB>
host.conf hosts hosts.allow hosts.deny httpd/
<bash>: ls /etc/h
The shell tells you what your choices are and is again ready for more input. This also works on commands
too:
<bash>: mo<TAB><TAB>
modemlights_applet montage mount.smbfs
modemtool more mouse-properties-capplet
modinfo morepgp mouse-test
modprobe mount mouseconfig
<bash>: mo
This shows us the possibilities, and again returns us so we can continue typing a command.
Remember that tcsh has the same possible-completion mechanism as bash does above. Just use <Ctrl-
D> instead.
3.6.3 Command Line Substitution and History

In both tcsh and bash we have the facility of Command Line Substitution and Command History.
Command Line Substitution allows you to substitute a dynamic expression inside a command:
<bash>: echo My name is $USER

My name is Joe
<bash>:
The $ symbol allows bash to process the expression in-line and later provide the results to the echo
command.
Command History allows you to use stuff from your old commands in your current command. Remember:
recycling is good for the environment ♣. The following list shows the history reference syntax common
to both bash and tcsh:
!! Redo the last command. Same as !-1
!−N Repeat the N th most recent command (see next)
!−3 Repeat the 3rd most recent command
!N Redo the N th entry in the history list
!string Redo last command starting with the text “string”
!?string? The most recent command which contains the text “string”
Figure 3.3: History Referencing Specifications
On top of history referencing, you can add these modifiers to the shell command line by appending them
to the history reference after a colon (:)

0 The first (command) word

n The nth word on the command line besides the command.
^ The first word, equivalent to ‘1’
$ The last word
% The word matched by an ?s? search
x-y A range of words. ‘-y’ abbreviates ‘0-y’.
* Equivalent to ‘^-$’, but returns nothing if
the event contains only 1 word
x* Equivalent to ‘x-$’
x- Equivalent to ‘x*’, but omitting the last word (‘$’)
Figure 3.4: History Modifiers
3.6.4 Exercises
You can get a list of your history by typing simply history. Create two files called boogie.man and
boogie.man.old. The file contents are not important for this exercise. Try these examples of the above
history machinery to try. The exact history will vary depending on what you do so please adapt these to
your current situation:
<bash>: history | tail -4

9 8:30 touch boogie.man
10 8:31 cp boogie.man boogie.man.old
11 8:36 echo hello >> boogie.man
12 8:37 diff boogie.man.old boogie.man
<bash>: !-2
vi boogie.man
<bash>: diff !-2:2.old !-2:2

diff boogie.man.old boogie.man
<bash>: diff !!:1.old !!:2

diff boogie.man.old.old boogie.man
diff: boogie.man.old.old: No such file or directory
<bash>: !?cat?:0 !!:2

cat boogie.man

Chapter 4
Text Files
4.1 Command Line Tools

Colaboration and Productivity Tools

cal [[month] year] Display calendar for month or year
date Display or set system date and time
w Display information on users currently logged on
who Display information on users currently logged on
whoami Display userid of current user
talk user@hostname Initiate a talk session with another user
write user@hostname Send a message to another user
msg y | n Enable, disable reception of talk and message from other
users
mail user@hostname ¡ file Send file as an email message to user@hostname
ispell file Look for misspellings in file and offer suggestions
aspell file Aspell (replacement for Ispell) looks for misspellings and
offers suggestions
look prefix Display words beginning with prefix
sort [rules] file Sort file according to rules
wc file List the number of characters, workds, and lines in file

CHAPTER 4. TEXT FILES
diff file1 file2 List the differences between file1 and file2
cat list of files Concatentate files and sent to standard output
more file Display file one page at a time
less file Display file one page at a time and allow paging backwords
head file Display the first few lines of a file
tail file Display the last few lines of file

word)
ronment.
Printing
ers.
Text files may require special handling to print them on a postscript printer. The command
cat textfile | a2ps | lpr
converts textfile to postscript and prints it on a postscript printer.
4.2 Vi and related editors

Although vi is considered one of the most primitive editors in Linux , it is still the most native of the
UNIX editors. In a pinch, you will have to know how to use vi if your favorite editor is not available.
vi and vim are screen oriented keyboard driven editors: they display the text one screen full at a time
and editing actions take place at the cursor. They are also modal editors; they work in two modes: a
command mode and an editing mode. There is also a line oriented editing mode in which editing actions
take place at designated locations. VI is case sensitive in both modes.

4.2. VI AND RELATED EDITORS
4.2.1 The basics

Interaction with vi is best thought of in terms of a finite state machine.
• viYourFile (launches the editor in command mode)

• insert mode: i or o (text is then entered by normal typing)
• return to command mode: ESC
• save file and exit vi: ZZ or :wq Return
• exit without saving: :q! Return
• save file and continue in editing: :w Return
On occaision other applications (mail, write, talk) will display text over an editing session. Type CTRL-l
to redisplay your editing session.
4.2.2 To enter vi
at the system prompt, enter
• vi fileName or
• vim fileName
where fileName is either a pre-existing file or the name of a new file. You are now in the
command mode.
4.2.3 To exit vi
press ESC to make sure you are in command mode
1. Type ZZ or :wq to write (save the file) and quit

2. Type :q or :q! to quit without saving the file
4.2.4 Insert text

(press ESC to make sure you are in command mode)
• enter i (aA - append; iI - insert; oO - open lines)

• enter the desired text
• press ESC to return to command mode

4.2.5 Move the cursor

• Use either the arrow keys or use

– h to move left one character
– j to move down a line
– k to move up a line
– l to move right one character
• Type H to move to the head of the file (top left)
• Type L to move to the end of the last line (bottom right)
• Type G to go to the beginning of the last line
• Type nG or :n to go the beginning of the n-th line from the current position
• Type w to go to the beginning of the next word
• Type b to go back one word
• Type 0 (zero) to go to the beginning of the current line
• Type $ to go to the end of the current line
• Type CTRL+u to go up one-half screen
• Type CTRL+d to go down one-half screen
• Type CTRL+f to go forward one screen
• Type CTRL+b to go back one screen
4.2.6 Delete text

n x to delete the character above the cursor; n is an optional number of charactors
• [n]dw to delete the current word

• [n]dd to delete n line(s)
• Type D or d$ to delete from the cursor to the end of the current line
• Type d0 to delete from the start of the line to the cursor

4.2. VI AND RELATED EDITORS
4.2.7 Additional commands

• Undo a command
– Type u to undo the previous command
• Search for a Pattern (press ESC to make sure you are in command mode)
– Type /pattern position the cursor at the first occurrence of the pattern following the cursor
– Type ?pattern to position the cursor at the first occurence of the pattern preceding the cursor.
– Type n to move the cursor to the next occurrence
• Seach and Replace (press ESC to make sure you are in command mode)
– Type :from,to s/target/replacement/g to replace all occurances of targetwith replacementfrom
line from to line to. The command :1,$s/target/replacement/g to replaces all occurances of
targetwith replacement.
• Save a file
– Type :w to write the file to storage
– Type :w fileName to write the file to storage as file fileName
• Add text from another file
– Type :r fileName to read the file fileName and insert it at the cursor
• Copying or Cutting and Pasting text
– Delete comands put deleted text into a buffer
– Type p, P to copy (put) text from buffer into editor at the cursor
• Join lines
– Type J to join lines
• Open lines
– Type O to open lines
vi has 2 basic modes: Command mode, and Insert mode. First we will start to work on our project.
Basic command mode functions are listed below. Those with and star beside them indicate commands
which enter insert mode.
These commands may not seem obvious from a mnemonic point of view, but were designed for simplicity
and keyboard efficiency.
You can read and write files in command mode by using ex mode which is invoked by typing : (which
will prompt you at the bottom of the vi screen)
:w Write this file out.

:q Quit vi w/o writing.
:q! Force quit vi w/o writing.
:w! Force write.
:wq Write this file AND quit vi.
:r file Read in ‘‘file’’ before cursor.

key Insert Description

A * Append after this line
a * Append after this character
C * Replace to end of line
cw * Replace this word
i * Insert (start typing here)
O * Open to the line above cursor
o * Open to the line below cursor
R * Overwrite Mode
r Overwrite with the next char
j Move cursor one line downward
k Move cursor one line up
l Move cursor one char right
h Move cursor one char left
x Delete character
w Move one word forward
b Move one word back
D Delete to end of line
dd Delete this line
dw Delete word
yy Yank a line into buffer
p Insert buffer lines after line
P Insert buffer before this line
<esc> Exit insert mode
Table 4.1: vi Command Mode Operations
Find, replace, and execute commands are done in a similar way:
/pattern F ind ‘‘pattern’’ and move cursor there

:1,100 s/blue/green/g F or line 1 to 100, substitute green for blue
:1,$ s/blue/green/g F or all lines, substitute green for blue
:’a,’b s/pie/cake/g F rom mark ‘‘a’’ to mark ‘‘b’’, substitute out pie
:1,$ !sort sort all lines
Once we get familiar with vi we will discuss these modes further. The first thing to do is to start the
editor by typing the following:
<bash>: vi input.txt
You should now have an empty screen that looks like
˜
˜
˜
Now type i and start typing some meaningful text like
Mary had a little lambda, who’s feet was white as snow.

˜<esc>
˜
˜

4.3. EDITING WITH PICO
The i puts you into insert mode, and the final <esc> puts you back into command mode. Now write
and quit by doing a :wq
Mary had a little lambda, whose fleets was white as snow.

˜
˜
˜
˜
:wq
"input.txt" 1 line, 58 characters written
4.3 Editing With Pico

Pico is by far the easiest editor to use since it usually displays its commands at the bottom of the screen.
Start by typing pico input.txt:
UW PICO(tm) 3.5 File: input.txt

Mary had a little foo, whose fleece was white as snow....
Start xxx Regular
bar baz foo them@nowhere.com Regular
Linux Rocks TurboLinux eats apples Respect
Some say Linux Rocks
^G Get Help Ô WriteOut ^R Read File ^Y Prev Pg ^K Cut Text ^C Cur Pos
^X Exit ^J Justify ^W Where is ^V Next Pg Û UnCut ^T To Spell
The only problem with pico (and other non-vi editors) is that it may not be on your system in an
emergency situation, which is when you need it the most. Hit ^X to exit pico.
4.4 Editing With emacs

emacs is the most powerful off all the editors commonly used under Linux . The X-Windows version
is fully loaded with all kinds of option menus and panels. Despite its complexity, you can get by with a
handful of commands. In this section, C- represents Ctrl-, so for example C-x is Ctrl-x.
emacs Exiting and File Manipulation

save a file back to disk C-x C-s
save all files C-x s
exit emacs permanently C-x C-c
read a file into emacs C-x C-f
insert contents of another file into this buffer C-x i
suspend emacs (or iconify it under X) C-z
goto beginning of line C-a
goto end of line C-e
delete character C-d
cut to end of line C-k

cut and buffer text C-w

set mark C-space
paste buffer C-y
emacs Error Recovery and Searching

undo an unwanted change C-x u
abort partially typed or executing command C-g
redraw garbaged screen C-l
search forward C-s
search backward C-r
regular expression search C-M-s
reverse regular expression search C-M-r
emacs Formatting
indent current line (mode-dependent) TAB
indent region (mode-dependent) C-M-
indent sexp (mode-dependent) C-M-q
indent region rigidly arg columns C-x TAB
emacs Help
basic help C-h
scroll help window C-M-v
emacs Tutorial C-h t
apropos: show commands matching a string C-h a
show the function a key runs C-h c
describe a function C-h f
4.5 Using Mail

Linux comes with several mail handlers. I prefer elm because I know it well and it is powerful, however
pine is very user friendly and easy to use. We will discuss both.
If the system you use provides email service but you do not wish to use the local service, create the file
.forward in your home directory containing the email address where you wish to receive your email.
4.5.1 Elm
Fire up elm by typing “elm”. When it first starts, it will ask you some questions about files, just say
yes to all.
Mailbox is ’/var/spool/mail/carinhas’ with 2 messages [ELM 2.5 PL1]
1 Jan 12 * Mail Delivery Subs (61) Returned mail: Host unknown

2 Jan 12 * Mail Delivery Subs (62) Returned mail: Host unknown
| =pipe, !=shell, ?=help, <n>=set current to n, /=search pattern

a)lias, C)opy, c)hange folder, d)elete, e)dit, f)orward, g)roup reply, m)ail,
n)ext, o)ptions, p)rint, q)uit, r)eply, s)ave, t)ag, u)ndelete, or e(x)it

4.6. REGULAR EXPRESSIONS
Command:q
Once inside elm, just hit “m” at the command prompt, and follow the signs. elm is more command
driven and UNIX -like. When in doubt, just keep hitting “q” for quit. Use the arrow keys to go up and
down the message list.
4.5.2 Pine
Pine is easier to learn and use, and is compatible with elm, so you can use them interchangeably. Use
the up-arrow, down-arrow, <, and > arrows to navigate.
PINE 4.10 MAIN MENU Folder: INBOX 2 Messages
? HELP - Get help using Pine

C COMPOSE MESSAGE - Compose and send a message
I MESSAGE INDEX - View messages in current folder
L FOLDER LIST - Select a folder to view
A ADDRESS BOOK - Update address book
S SETUP - Configure Pine Options
Q QUIT - Leave the Pine program
? Help P PrevCmd R RelNotes

O OTHER CMDS > [ListFldrs] N NextCmd K KBLock
4.6 Regular Expressions

Regular expressions as used in vi, grep, lex, flex etc. Regular expressions are formed from single characters,
concatenation of regular expressions, choice between two regular expressions, and zero or more occurences
of a regular expression.
RE ::= c | RE RE | RE + RE | RE*
UNIX extends regular expressions with additional metasymbols and extensions to the regular expression
notation. Regular Expressions are search patterns that give you control over how text patterns are found.
Here are the basics pattern matching symbols:
. Match any single character. ‘‘.*’’ matches anything any number of times. See *
* Match the last char or exp 0 or more times. a* matches (), a, aa, aaa,
+ Match the last char or exp 1 or more times. a+ matches a, aa, aaa, ..
? Match the last char or exp 0 or 1 time. a? matches (), a
^ Match the beginning of the line. âbc matches lines starting with abc.
$ Match the end of a line. ^$ Matches empty lines.
[] Match chars inside. [abc] can match a, b, or c or in a range [A-Z].
[^] Negates the match chars: [âbc] matches anything except a, b, or c.
() Char grouping. (abc)+ matches abc, abcabc, abcabcabc, ....
\ Match regex chars like \+, \*, and \^.
{n} The preceding item is matched exactly n times.
{n,} The preceding item is matched n or more times.

{,m} The preceding item is optional and is matched at most m times.

{n,m} The preceding is matched at least n times, not more than m times.
\b Match a word boundary: abc\b matches abc but not abcd.
\B Match a non-boundary: abc\B matches abcd but not abc.
\w Match a word: abc\w matches abcdef but not abc@.
\W Match a non-word: abc\W matches abc& but not abc.
4.6.1 GREP
grep is a pattern search tool. It will find patterns in a file and print it on the terminal. These patterns
are regular expressions as well. Test grep with the file as before.
grep [ options ] pattern file..
# 1. Find lines with ‘‘Linux’’ ANYWHERE in the file.

grep Linux input.txt
# 2. Find ‘‘Linux’’ in a non word boundary like ‘‘TurboLinux’’

grep ‘‘\BLinux’’ input.txt # Remember \B?
# 3. Find ‘‘Linux’’ at beginning and the end of line, regular expressions.

grep ’Linux$’ input.txt # Note the single quotes to trick the shell,
grep ^Linux input.txt # because shell tries to interpret stuff like $.
# 4. Lines start with ‘‘Linux’’, end with ‘‘Respect’’.

grep ’^Linux.*Respect$’ input.txt # ‘‘.*’’ means any number of anything.
# 5. Find some tough patterns:

grep ’<.*>’ input.txt # Find lines that have ‘‘<something>’’. Html tags?
4.6.2 AWK
Awk is a pattern scanning and process language. Its Syntax is
awk ’program’ input-files
When the program is long, it is usually more convenient to put it in a file and run it with a command
like this:
awk -f program-file input-files
Take a look at the manual now. Here are some one-liner examples:
awk ’/xxx/’ input.txt # Find ‘‘xxx’’ in input.txt

Most all regular expression syntax works inside the //’s.
awk ’NF > 0’ input.txt
This program prints every line that has at least one field. Its an
easy way to delete blank lines from a file.
awk -F: ’{ print $1 }’ /etc/passwd | sort

4.6. REGULAR EXPRESSIONS
This program prints a sorted list of the login names of all users.
awk ’END { print NR }’ input.txt
This program counts lines in input.txt.
awk ’NR % 2 == 0’ input.txt
This program prints the even numbered lines of input.txt.
If you’d use the expression ‘NR % 2 == 1’, it’d print odd numbered lines.
awk ’{if (length($0) > max) max=length($0)} END {print max}’ input.txt
This program prints the length of the longest input line.
awk ’length($0) > 80’ input.txt # Print lines longer than 80 characters.
Since the sole rule has a relational expression as its pattern,
and has no action, the default action, printing the record, is used.
awk ’BEGIN { for (i = 1; i <= 7; i++) print int(101 * rand()) }’
This program prints seven random numbers from zero to 100, inclusive.
ls -lg *.txt | awk ’{ x += $5 } ; END {print "total bytes:" x }’
This program prints the total number of bytes used by all .txt files.
4.6.3 Perl
Perl, “Practical Extraction and Report Language” is the Swiss Army Knife of all scripting languages.
It has everything except a Megawatt laser, and a kitchen sink. Perl has a very complete set of regular
expressions and a large set of UNIX system calls that make it a steroid-pumped tool:
perl [options] [ program ]

On the command line you can use the syntax
perl -ne ’command’ file

where the options are
-e ’command’ execute command instead of script-file

-n run the script on each line of input file
-w Print WARNINGS for your script file
Even though complex, we can learn a few perl one liners that work miracles:
# 1. Find lines starting with ‘‘Linux Rocks’’ in input.txt.

perl -ne ’if ( /^Linux Rocks/ ) {print} ’ input.txt
# 2. Find ‘‘Linux’’ in a non word boundary like ‘‘TurboLinux’’

perl -ne ’if ( /\BLinux/ ) {print} ’ input.txt
# 3. Find ‘‘Linux Rocks’’ and replace it with ‘‘Rocks Linux’’.

# Regular expressions matching ()’s get named $1, $2, $3, etc..
perl -ne ’s/^(Linux) (Rocks)/$2 $1/; print ’ input.txt
# 4. Lines start with ‘‘Linux’’, end with ‘‘Rocks’’.

perl -ne ’print if /^Linux Rocks$/ ’ input.txt
# 5. Find email addresses in input.txt:

perl -ne ’print if (/(\w[.\w]*\@\w[\w_-]+\w)/gm)’ input.txt

4.6.4 SED
sed is a stream editor, which means that it works like a conveyer belt on your terminal. The text flies
past you while it works.
sed [script ] file..
sed is complex so we’ll just show some simple usage and examples:
# 1. From line 1 to 32, print the line number

sed ’1,32 {= ; }’ input.txt # Almost works in VI
# 2. Delete lines 1 through 32

sed "1,32 d" input.txt # :1,32 d in VI
# 3. Under UNIX: convert DOS newlines (CR/LF) to Unix format

sed ’s/^M$//’ input.txt # In bash ^M = C-v C-m, tcsh ^M = C-v C-j
# 4. Under DOS: convert Unix newlines (LF) to DOS format

sed ’s/$//’ input.txt # method 1
sed -n p input.txt # method 2
# 5. Delete leading whitespace (spaces/tabs) from front of each line

# ’^t’ represents a true tab character. ^t => Ctrl-V Ctrl-I.
sed ’s/^[ ^t]*//’ input.txt # OK in VI without single quotes.
# 6. Delete trailing whitespace (spaces/tabs) from end of each line

sed ’s/[ ^t]*$//’ input.txt # see note on ’^t’, above
# 7. Delete BOTH leading and trailing whitespace from each line

sed ’s/^[ ^t]*//;s/[ ^t]*$//’ input.txt # see note on ’^t’, above
# 8. Substitute "foo" with "bar"

sed ’s/foo/bar/’ input.txt # replaces 1st instance foo with bar.
sed ’s/foo/bar/4’ input.txt # replaces only 4th instance in a line
sed ’s/foo/bar/g’ input.txt # replaces ALL instances within a line
sed ’/baz/s/foo/bar/g’ input.txt # Sub ONLY on lines which contain "baz"
# 9. Match lines that have ‘‘Regular’’ in them:

sed ’/^Regular/d’ input.txt # Delete lines starting with ‘‘Regular’’
sed ’/Regular$/d’ input.txt # Delete lines ending with ‘‘Regular’’
#10. From ‘‘Start’’ to ‘‘Linux’’, substitute apples for oranges

sed /Start/,/Linux/s/apples/oranges/ input.txt

Chapter 5
The Linux Environment
In Linux and UNIX , you carry many local attributes that control how your immediate shell
interface interacts with the system. These attributes are called Shell and Environment Vari-
ables, and they are critical to Linux operation. UNIX and Linux use many variables for internal
processing. These help the system manage resources for users. This chapter discusses the bare
necessities that all Linux survivalists need to know.
Personal Information and Environment

.login File in home directory which Personalize the csh and tcsh environment. Read upon lo
.cshrc Files in home directory which personalize the csh, tcsh environment. Read upon login
.profile File in home directory which personalizes the sh, bash environment
.forward File in user home directory containing email forwarding information
.signature Automantically added to your email
.plan File in user home directory containing public plans
.project File in user home directory
.nofinger If present denies existence of user outside of local host. Created by touch .nofinger
finger user[@RemoteHost] Lookup user information contained in password file and
the following files.
chsh Change user’s login shell
chfn Change use name and information in /etc/passwd file
source file Update environmental information from file (.cshrc).
5.1 The UNIX Shell Game

One prominent feature of UNIX is that you can control the interface by which you communicate with it.
These interfaces are called shells and there are too many. The original shell is called sh and it has many
close cousins including ash, bash, and ksh. bash is the “native” Linux shell and is the default shell
for Linux . Thus, bash is used for almost all startup files (scripts) and maintenance scripts for Linux .
Many users prefer to use the Berkeley UNIX C-type shells like csh and tcsh. Your choice is largely a
matter of preference, however, we will use bash as the default in the course. We will also discuss tcsh
because of its easy to understand C-like syntax.
Normally a shell can be invoked anywhere in your current login session simply by typing the shell name.
This is useful for testing, but normally, you will use your default shell.

CHAPTER 5. THE LINUX ENVIRONMENT
A user can control fundamental aspects of his/her environment by setting shell variables as you work,
or inside of a startup script file which gets invoked when you login to your account. Simple examples
common to each shell are your prompt (the thing that stares at you and waits for your commands) and
aliases. An alias is a shorthand for some other command which a user can define, normally in the shell
startup files like .bashrc and .cshrc. We’ll work with these concepts very shortly.
5.1.1 Shell Features

Most Linux distributions come with bash and tcsh which replace sh and csh respectively. These are
the two big shells that 99.9% of the Linux world uses. There are several key commonalities and differences
between the two. The following tables outline these:
Action bash tcsh

Default Prompt $ #
Force Redirection >| >!
Force Append >>!
Variable assignment var=value set var=value
Set Environment var export var setenv VAR value
Number of arguments $# $# ($#argv in csh)
End loop statement done end
End case statement esac endsw
Loop for/do foreach
If statement if [ ... ] if ( ... )
End if fi endif
Read from terminal read $<
Start until loop until/do until
While loop while/do while
Table 5.1: Bash-Tcsh Difference
5.2 Bash
Bash is the default shell for Linux . The default shell is like an environment where all the basic commands
come from bash. Let’s look at the bash manual page:
<bash>: man bash

Look at the default prompt on your screen. Notice that the prompt contains the shell name and username.
The prompt in bash is controlled by the variable PS1. Type in the command “set” to get the following
sort of output:
PRINTER=astro
PS1=<\s>:
PS2=>
PS4=+
PWD=/home/joe
SHELL=/bin/bash
SHLVL=2
TERM=vt100

5.2. BASH
Symbol/Command Action
> Redirect output
>& Redirect stdout and stderr
>> Append output to file
| Pipe output
& Run process in background
TAB Filename/command completion
!n Repeat command #n in history
$var Variable substitution
# Comment line
bg Put process in background
fg Put process in foreground
cd Change directory
echo Echo execution output
jobs List current shell processes
kill Kill specified processes
umask Set default file permissions
Table 5.2: Bash-Tcsh Commonality List (Incomplete)
5.2.1 Setting Your Prompt

Note that set provides the definition for PS1 and several other variables. Lets look up the manual on
bash by typing “man bash”. You are looking at an endless stream of information from the native Linux
manual pages. Type “/prompt” to search for the word “prompt”. You can find the next occurrence by
typing “n”. Keep going until you see something similar to
PROMPTING
When executing interactively, bash displays the primary
prompt PS1 when it is ready to read a command, and the
secondary prompt PS2 when it needs more input to complete
a command.
\t the current time in HH:MM:SS format

\d the date in "Weekday Month Date" format
\s the name of the shell, the basename of $0
\w the current working directory
\W the basename of the current working directory
\u the username of the current user
\h the hostname
\# the command number of this command
\! the history number of this command
\$ if the effective UID is 0, a #, otherwise a $
Lets change the bash prompt by adding the command number. This is an example:
<bash>: PS1=’ \s: \u \t:# \#: ’
which should produce something like

bash: joe user 00:02:07:# 13:
Now lets set your prompt so that it always displays your path. Even though this manual will specify
<bash>: , you should still use a prompt like this:
<bash>: PS1=’<\w>: ’
5.2.2 Creating Aliases (Shortcuts)

Now as it happens, I get very tired of typing out ls -Agl every time I want a long directory listing, so
I want to alias that command to d ( for “directory”) as follows:
<bash>: alias d=’ls -Agl’

We really would like to make this definition permanent, by putting this command into .bashrc, and we
will learn how to do this in Chapter 10 on scripts.
The same alias in tcsh (or csh) is written without the equals (=) sign: You start tcsh by typing
tcsh:
<bash>: tcsh
<tcsh>: alias d ’ls -Agl’
<tcsh>: d
.............
<tcsh>: exit
5.3 Shell Variables

In Linux , you always keep shell variables user, directory, path, group, and many others. These variables
help your shell cope with all the information it needs to deal with normal operations. As we discussed
before, the set, env, printenv commands will display your environment variables: (without #comments)
PWD = /home/carinhas/misc # Your current working directory

GID = 100 # Your current group ID number
GROUP = users # Your current group name
HOME = /home/joe user # Your home directory
HOST = astro # The name of your workstation.
PATH = /bin:/usr/bin # Your current path to executable commands.
PS1 = <\s>: # Your current prompt
SHELL = /bin/bash # Your current shell
UID = 501 # Your user ID number
USER = carinhas # Your user name
Your current working directory (cwd) is where all files are sought and written by default. uid and gid are
the numerical values for your user and group, which identify you as a user and group member respectively.
Your home is where you (and your files) live, and your path is where you look for your prey (executable
commands).
You can always determine your cwd with the pwd command. Your cwd is always referred by the “dot”
(.) symbol, and the directory directly below is referred to by the double-dot (..) symbol as in the result
of an ls -Agl:

5.3. SHELL VARIABLES
<bash>: ls -gl
total 608
drwxr-xr-x 2 joe user users 1024 Dec 20 19:09 .
drwxr-xr-x 15 root root 1024 Dec 10 19:04 ..
-rwx------ 1 joe user users 3185 Dec 19 15:51 .cshrc
-rwx------ 1 joe user users 3185 Dec 19 15:51 .bashrc
-rw-r--r-- 1 joe user users 33 Dec 20 10:49 doubletake
drwxrwxrwx 1 joe user users 33 Dec 20 10:49 mail
Note that “..” refers to /home when you are in your home directory /home/joe user. The first character
in the long file listing is the file type, which tell us what kind of animal we have. Also ∼ refers to your
home directory in your prompt, and changes to the /fully/qualified/path only when moved out of
/home/joe user .
5.3.1 Environment Variables

Environment variables are shell variables that control the so-called ’environment’, or shell options. Envi-
ronment variables are used internally by the shell at all times. Normally, user defined shell variables will
not affect how the shell works.
Environment variables are all uppercase words like PATH, SHELL, TERM, PAGER, PS1, MANPATH.
In bash, you set environment vars with the export:
NAME=value
export NAME
Or you can do it all in one command:
export NAME=value
Examples in bash are
<bash>: export PATH=/bin:/sbin:/usr/sbin

<bash>: DISPLAY=groucho:0
<bash>: EDITOR=/bin/vi
<bash>: export DISPLAY EDITOR
Environment variables are set in tcsh in the following way:
setenv NAME value
5.3.2 Exercises
Exercise 1
Find the hostname of your machine and replace “groucho” in the DISPLAY variable as above.

Exercise 2
Go into the tcsh shell and work the following examples:
<bash>: tcsh
<tcsh>: setenv PATH /bin:/sbin:/usr/bin:/usr/sbin
<tcsh>: setenv DISPLAY groucho:0
<tcsh>: setenv EDITOR /bin/vi
<tcsh>: env
5.4 Choosing the Right Path

Your PATH is used by the system to find commands. Its important to set this correctly so that you can
access everything that you need.
To set you path in bash we edit the .bash profile file:
<bash>: pico .bash profile

PATH=$PATH:$HOME/bin # Add $HOME/bin to $PATH
.... write and quit .... (hit Ctrl-X, type Y, return)
<bash>: source .bash profile # Initialize $PATH.
Once this is done, bash will look in your own /home/user/bin folder to find any commands you put
there. This is a good thing which will come in handy later when we work on scripting.
5.4.1 Exercises
Set your path in tcsh by editing your .login file, and add ∼/bin to your path.
<tcsh>: pico .login

set path = (/bin /sbin ~/bin /usr/bin /usr/ucb /usr/bsd ) # added ~/bin
.... write and quit ....
<tcsh>: source .login # Install the new $path.
<tcsh>: rehash # Tell csh to find files in $path.
tcsh uses its rehash function to re-read all the folders in your path and find the new commands.
5.5 Groups and Newgrp

Groups allow users to have access to files and programs jointly. This is what the group attribute is for.
Users can belong to many groups. You can belong to more than one group, however you can only be
active in one group at a time. All users have a default group that they enter at login. You change your
group interactively by using the newgrp command:
<bash>: newgrp groupname

5.5. GROUPS AND NEWGRP
which will spawn a new shell (on top of the old one) with your default group given by groupname. For
this trick to work, you must belong to this group to begin with.
Once you are in this new group state, all process and files created will carry the group name as well as
your user name. Having multiple groups can be useful when accessing restricted hardware or working in
a team of developers, where everyone needs access to common software and systems.
To exit this group-shell, simply type exit.
5.5.1 Exercises
1. Look at the files /etc/passwd and /etc/group. Determine how groups are created and assigned
with the help of the manual pages. Write your observations down here.


Chapter 6
User Level Security
Protection and Security

.nofinger File in users home directory created using the touch com-
mand limits finger to local machine
passwd Change user password
umask Set default file access permissions (permissions taken away
chmod [ugo][+|-][rwx] file Change file mode (permissions)
chmod octalmode Change file mode (permissions)
ssh [user@]RemoteHost[command] Secure shell, replacement for telnet
sftp RemoteHost Secure file transport, replacement for ftp
vim -x file Create encrypted file. You will be promped for an
encryption key
Examples:
• ls -l
• Protection bits: u - user, g - group, o - other.
[-dl] rwx rwx rwx
• chomd mode filename
chmod who op permissions filename
Examples: chmod go-w ..., chmod o+x ..., etc
• umask - set file creation mask (ppermissions taken away; in .login file.
• chown – must be performed by root
• ln -s q sl – create a symbolic link sl in working directory to file/directory q
6.1 Security
The primary concern is integrity. Accuracy or validity is fundamental and security is secondary. Security
is protection against unauthorized disclosure, alteration, or destruction of data and protection against
unauthorized use.

CHAPTER 6. USER LEVEL SECURITY
6.2 Social Engineering

passwords ... system access ...
6.3 Cryptography
6.3.1 SSH, SCP, and SFTP
6.4 Authentication and Integrity

6.4.1 Digital Signatures
6.4.2 Cryptographic Checksum
6.5 Privacy
6.5.1 Privacy-Enhanced Electronic Mail (PEM)

Chapter 7
Filesystem Essentials
In this chapter, we will learn how to master files and filesystems. This is a very important
aspect of Linux , and is used constantly.
Compression and Archives:

compress file file.Z Compress a file
uncompress file.Z file Uncompress a file
gzip file file.gz Compress a file
gunzip file.gz file Compress a file
tar Tape archive untility
tar cf Archive.tar Dir Create an archive (Archive.tar) of the directory (Dir) and
its contents.
tar cfz Archive.tar Dir Create a compressed archive (Archive.tar) of the directory
(Dir) and its contents.
tar xf Archive.tar Dir Extract the tar file into the working directory.
tar xzf Archive.tgz Dir Extract the compressed tar file into the working directory.
tar tzf Archive.tgz Dir List the contents of the tar file.
tar xf Archive.tgz FileofInterest List the contents of the tar file.
File mode and ownership:

chmod Change file mode
chown Change file ownership or group
7.1 The Linux Virtual Filesystem (VFS)

All files in Linux are accessible from a single integrated filesystem called a Virtual Filesystem. It is a
tree structured filesystem with directories and files. The base of that system is called the root and often
denoted by a slash (/). Users home directory is found in /home. User with user name joe will have
/home/maria as The mc is a visual shell for use in a command line environment and it may be used to
navigate around the system files below:
The next figure outlines the root VFS root tree as set by the Filesystem Hierarchy Standard located at
http://www.pathname.com/fhs/
Among the most important of these are /, /etc, /usr, and /opt, /tmp and /home. Below is a sample
directory tree that starts from root but lacks the /var subtree.

CHAPTER 7. FILESYSTEM ESSENTIALS
Figure 7.1: Linux VFS as viewed by mc
The root directory

bin Essential command binaries
boot Static files of the boot loader
dev Device files
etc Host−specific system configuration
lib Essential shared libraries and kernel modules
mnt Mount point for mounting temporary filesystems
opt Add−on application software
sbin Essential system binaries
tmp Temporary files
usr Secondary hierarch
var Variable data
Figure 7.2: VFS Root Tree
7.1.1 Exercises
1. The tree diagram above is not correct for a modern version of Linux. Find and compare the
differences with your system and write down your results.
2. The /proc filesystem is a special on that has direct access to the system kernel. The files in /proc
have no size, but what happens if you do the following:
<bash>: cat /proc/cpuinfo
<bash>: cat /proc/pci

7.2. FILE ATTRIBUTES
/ bin
dev
etc
home mary
jane
lib joe
proc
tmp
usr X11R6
bin
etc
include
lib
local bin
man emacs
spool etc
lib
src linux
tmp
Figure 7.3: A Linux directory tree.
Discuss the possible uses of this.
7.2 File Attributes

Lets take a close look at a directory listing of mail. Enter the command ls -agl mail:
drwxrwxrwx 1 joe user users 33 Dec 20 10:49 mail

-file type
-user read
-user write
-user execute
-group read
-group write
-group execute
-other read
-other write
-other execute
link number
user ID
group ID
Size
Month
Day
Time
File Name
Figure 7.4: Unix File Attributes
Its entry gives us a lot of information. It has all of its access entirely open, so everyone can read, write,
and execute. Also note that the first attribute listed is its type, in this case a “d” for directory. File
attributes for restricted (plain type) files are listed as dashs (-).
The next 9 parameters are file permissions or access parameters In UNIX , files have 3 sets of 3 access
parameters or “permissions”. The first 3 are for the owner, the second 3 for the group, and the last 3 for
“other” or the “world”. The permissions are read (r), write (w), and execute (x). When you look at a file

listing in long format you will see these attributes listed as (r,w,x) or omitted with a “-” in their place.
If the attributes are present, access is granted to for that action, otherwise access is denied.
Other major file types include symbolic links (l), an internal reference to another file, and plain files (-).
The following table summarizes native Linux file types as listed by the find command. Note that the
plain file permission never lists by ls as (f) but as a dash (-).
b block special file

c character special file
d directory
l symbolic link
p named pipe (FIFO)
s socket
Table 7.1: File Types as Listed by find
7.3 Changing File Attributes with chmod

An important task for any Linux user is to manage file attributes (permissions) with the chmod command.
The syntax for chmod is
chmod [-R] MODE FILE
where MODE can be a comma separated combinations of 3 sets of symbols:
[ugoa][+ - =][rwx]
u --> user
g --> group
o --> others
a --> all
You can add (+), subtract (–), or exactly set (=) the mode. Examples:
chmod u+w mail # Add user (u) write (w) privileges

chmod u-w mail # Remove user (u) write privileges.
chmod g+w mail # Add group (g) write privileges
chmod g=r mail # Allow only the group read privileges, nothing else.
chmod o+x mail # Add execute (x) privileges for others.
chmod a+x mail # Add execute (x) privileges for everyone..
chmod a=xr mail # Allow read and execute only to everyone..
chmod go-r mail # Remove group and others read privileges..
chmod ugo+w mail # Same as chmod a+w mail
MODE can also be an octal representation which represents the absolute mode:

7.3. CHANGING FILE ATTRIBUTES WITH CHMOD
1 = others execute # o=x

2 = others write # o=w
4 = others read # o=r
10 = group execute # g=x
20 = group write # g=w
40 = group read # g=r
100 = user execute # u=x
200 = user write # u=w
400 = user read # u=r
The trick with octal is that you can add up the different numbers to get all different combinations of
privileges like:
1 = others execute # o=x

+ 2 = others write # o=w
+ 4 = others read # o=r
-----------------------------
= 7 = other read, write, and execute = o+rwx
10 = group execute # g=x

20 = group write # g=w
00 = no group read #
100 = user execute # u=x
000 = no user write #
400 = user read # u=r
-------------------------- add these last 6 to get
530 = u=rx, g=wx
chmod 750 mail => chmod u=rwx,g=rx,u= mail

chmod 700 mail => chmod u=rwx,g=,u= mail
chmod 751 mail => chmod u=rwx,g=rx,o=x mail
chmod 741 mail => chmod u=rwx,g=r,o=x mail
7.3.1 Exercises
1. On directories, the execution privilege allows for access into the directory, so having just read and
write (rw) privilege is not enough. Only execute privilege will allow any access to anything inside
the directory. Try this out by doing a
<bash>: chmod 600 mail

<bash>: cd mail
There are actually many other modes that can be set using the first (non-octal) method. The ones
we discussed are the essential ones. Please check out the man pages on chmod and ls.
2. Restore your mail directory back to a usable mode.
3. Write down the command you just used in both octal and symbolic modes.
4. Read the man pages and learn how to set and the function of the “set user ID on execution” (SUID)
bit for chmod. Discuss this with your classmates and instructor. This concept is a requirement for
the LPI exams.

7.4 Changing File Ownership with chown and chgrp

Just as you can change the file permission, you can change the owner and group of a file. chown does
this for you with the syntax:
chown [-R] USER[.[GROUP]] FILE
Note that chgrp can change the group at the same time as in the following examples.
<bash>: chown johnson mail # Sorry, only johnson or Root can do that.
chown: mail: Operation not permitted
<Super>: chown joe user.users mail # Change owner to joe user and group to users.
<Super>: chown joe user.bozos mail # Change owner to joe user and group to bozos.
chown: mail: Operation not permitted # You’re not a member of bozos group!
<Super>: chown .testers mail # Ok, you are in the testers group.
chgrp is less powerful and not really needed but should be noted. Its syntax is
chgrp [-R] GROUP FILE
Here is an example.
<bash>: chgrp testers mail # Same as ‘‘chown .testers mail’’

<bash>: chgrp users mail # Same as ‘‘chown .users mail’’
7.4.1 Questions
Why can’t you change ownership of someone elses files? Who can do this? Is this a good or bad feature?
7.4.2 Exercises
1. Log in as root and change the owner and group of some files in the /tmp directory. Make sure you are
in /tmp and not /home/you/tmp or some other place.
7.5 Devices
In Unix (almost) all commands and hardware are represented by files. The hardware devices are fun-
damentally different and live in the directory /dev. Device files allow access to the hardware directly
through the kernel.
These special files have major and minor numbers that identify them in the Linux kernel. Major Device
Numbers specify a particular driver for I/O redirection in the kernel, and Minor Device Numbers specify
a device to be accessed by that driver. A typical example is the /dev/hda hard-drive:
brw-rw---- 1 root disk 3, 0 May 5 1998 /dev/hda

7.5. DEVICES
It has a major number 3 and minor number 0. The leading “b” indicates that this is a block device. It’s
owner is root and its group is disk .
Do an ls -agl /dev to see the major and minor numbers of typical devices.
We list the most commonly used devices below. Note that N will represent an integer number.
Device Name Type Description

/dev/hda B first IDE hard disk
/dev/hdaN B N th partition of /dev/hda
/dev/sda B first SCSI hard disk
/dev/sdaN B N th partition of /dev/sda
/dev/loop0 B First loopback device
/dev/console C System console
/dev/parport0 C first parallel port
/dev/lp0 C first printer port (LPT1)
/dev/null C System Trash Can (Black Hole)
/dev/ttyN C N th virtual console
/dev/pts/N C N th pseudo terminal (dynamically created)
/dev/ttySN C N th serial port
/dev/psaux C PS/2-style mouse port
/dev/mouse L Soft link to a serial mouse like psaux
/dev/printer S lpd local socket
Table 7.2: Linux Devices (See 7.1 for type info)
/dev/null is a funny device that swallows everything you feed it like a black hole, so don’t put anything
there that you want back. It’s very convenient for swallowing up unwanted error messages like this:
<bash>: netscape >& /dev/null &
We will learn the meaning of the preceding example in the next section. The cat command prints a text
file to the console and the cp command copies one file to another. Try this example
<bash>: cat .cshrc

<bash>: cp .cshrc /dev/null
<bash>: cat /dev/null
The last command should produce nothing.

7.5.1 Exercise: Redirect the output of command top to file top.log
7.5.2 Exercise:
<bash>: ls -l /dev | grep ˆc
<bash>: ls -l /dev | grep ˆb
<bash>: ls -l /dev | grep ˆl
<bash>: ls -l /dev/hda*
<bash>: ls -l /dev/sda*
7.5.3 Discussion

Chapter 8
Managing Filesystems
Creating and mounting filesystems is essential for disk access. The following sections discuss
the essential elements. Most of the commands we use are only accessible by the SuperUser
(root), so you must be logged in as such.
Filesystem Management
du
df
dd
fdisk
mount
sync
umount
mkfs
mkswap
swapon
8.1 Creating Partitions with Fdisk

On any new hard disk, you must first create the partitions that will be mounted. This is done with the
fdisk utility. The following is a typical fdisk application to add a 100MB partition to /dev/hdc:
8.1.1 Primary Partitions
<Super>: fdisk /dev/hdc

Command (m for help): n (create a new partition with n)
First cylinder (768-788, default 768):
Using default value 768
Last cylinder or +size or +sizeM or +sizeK (768-788, default 788): +100M
Command (m for help): p (print the current partition table)
Disk /dev/hdc: 255 heads, 63 sectors, 788 cylinders

Units = cylinders of 16065 * 512 bytes

CHAPTER 8. MANAGING FILESYSTEMS
Device Boot Start End Blocks Id System

/dev/hdc1 * 1 128 1028128+ 6 FAT16
/dev/hdc4 742 788 377527+ 5 Extended
/dev/hdc7 768 780 104391 83 Linux
8.1.2 Swap Partitions

A swap partition is just extra memory that lives on your hard drive. Most systems need swap if your
system has only a moderate amount of RAM.
To create a swap partition, toggle the Partition Id field to 82 with the “t” command while in fdisk.
We assume that we are still in fdisk from the previous section:
Command (m for help): t

Partition number(1-7): 7
Hex code (type L to list codes):82
Command (m for help): p

Disk /dev/hdc: 255 heads, 63 sectors, 788 cylinders
Units = cylinders of 16065 * 512 bytes
Device Boot Start End Blocks Id System

/dev/hdc1 * 1 128 1028128+ 6 FAT16
/dev/hdc4 742 788 377527+ 5 Extended
/dev/hdc7 768 780 104391 82 Linux Swap
Command (m for help): w (write the new partition info and exit)
The partition table has been altered!
8.2 Creating a Filesystem

After you have created a partition as in 8.1 you must first make the filesystem on it with mkfs (pronounced
“make fs”).
For example,
<Super>: mkfs -t ext2 /dev/hda7
This will write the inodes and the superblocks that the filesystem requires. Inodes are points on the
filesystem that correspond to filenames and storage. Superblocks are blocks of data that have critical
information about the filesystem, which can be used to fix the filesystem in case of a crash.
Lets create an ext2 filesystem on a floppy for fun. First use fdformat -n /dev/fd0. Then use mkfs
as ...
<Super>: mkfs -t ext2 /dev/fd0 1440

8.3. MOUNTING FILESYSTEMS BY HAND
8.3 Mounting Filesystems by Hand

A filesystem is unusable by Linux until you “attach” it to the operating system. This attachment is
called mounting. You can mount a filesystem with the mount command.
In order to mount a filesystem, you need to know the partition name (as in /dev/fd0) and a mount
point. A mount point is an empty directory where the filesystem will be located. It MUST be an empty
directory. Then you execute:
<Super>: mount -t ext2 /dev/fd0 /alt
Where the -t ext2 tells mount that the filesystem type is an ext2 filesystem, the default Linux filetype.
/alt is the mount point and /dev/fd0 is the device to be mounted. Go into /alt and see if you can
write on that partition.
To remove a filesystem, use the umount command:
<Super>: umount /alt
8.4 Swap Partitions

Swap partitions are an important source of virtual memory for the system. When you use all of your
RAM, Linux uses the Swap space for temporary storage.
8.4.1 Preparing Swap Space

Once your swap partitions are created, you need to prepare the swap partition with the mkswap com-
mand, which formats the partition:
<Super>: mkswap [-c] [-vN] [-f ] device [size]
The device can be either a partition or a file. Consult the man page now for mkswap options. A swap
partition is prepared in the following way:
<Super>: mkswap -v1 /dev/hda7
To create a swap file, you need to use the dd command as follows:
<Super>: dd -f=/dev/zero 0f=/swap bs=1024 count=8192

<Super>: mkswap -c /swap 8192
<Super>: sync
which will make a swap file of 8192 blocks (8MB) from /dev/zero and put it in the file /swap in your
root directory.
We used the sync command after mkswap to force the format information to be written to the filesystem,
otherwise the system may take its sweet time to do so. sync ensures that /swap gets written before a
crash or system failure.

CHAPTER 8. MANAGING FILESYSTEMS
8.4.2 Activating Swap Space

To activate a swap partition, use the swapon command:
<Super>: swapon /dev/hda7
To activate a swap file, use the following syntax:
<Super>: swapon /swap
where /swap is the swap file created as above.

To remove swap from the system, swapoff is used:
<Super>: swapoff /swap
8.5 Automatic Mounting at Boot with /etc/fstab

The /etc/fstab contains mounting information that the system uses at boot time. The format for
/etc/fstab is
# /etc/fstab
# device Mount Point Type Options Freq Pass
#
/dev/hda1 / ext2 defaults 1 1
/dev/hda2 /usr ext2 defaults 1 2
/dev/hda3 /home ext2 defaults 1 2
/dev/hda7 none swap defaults 0 0
/dev/cdrom /mnt/cdrom iso9660 noauto,ro 0 0
Edit this file now and add an entry for our hand-mounted filesystem for /dev/fd0 above. Once this is
done, you can abbreviate the mount command for it by simply doing
<Super>: mount /alt
Mount will look inside of /etc/fstab and get all the required information that it needs to do its work.
You can override /etc/fstab by specifying all the information as before.
Linux will automatically mount all the files listed correctly in /etc/fstab at boot time.

Chapter 9
Process Control
In this chapter we will learn how to control your processes both interactively and non-
interactively. It is very important to maintain control of your system and manage unruly
processes.
The program design philosophy under Unix is to treat each program as a filter which takes its
input from the standard input file and sends its output to the standard output. By default,
the standard input file is the keyboard and the standard output file is the display screen.
Programs can be strung together using pipes. A pipe takes the output of one program and
redirects it to become the input for another program.
IO Redirection, Pipes and Filters

cmd < file Use file as input to cmd.
Example: mail user@RemoteHost< file
cmd > file Create file containing the output of cmd.
Example: cat file1 > file2 ; equivalent to cp file1 file2
cmd >> file Append output of cmd to file file
cmd1 | cmd2 Connect two processes with a pipe. The output of the
first is piped to the input of the second
Processes and Multitasking

C-c Abort current task
C-z Suspend current task
bg Place suspened task in background
jobs List currentjobs processes
fg [JobId] Bring background task or suspeneded to foreground
task & Launch task into background
ps aux Report process status for all processes
kill -9 Pids Abort processes with Pids
top Display and update top CPU processes
nice Run a task with a modified scheduling priority
renice Alter priority of a running task
nohup cmd Run cmd as a background task immune to hangups.
crontab -e user Edit user’s cron file to schedule background tasks

CHAPTER 9. PROCESS CONTROL
9.1 Creating Foreground and Background Processes

Process control refers to the ability to control processes both in the foreground and background of the
shell. A foreground process is one that is running interactively in your shell, and include most processes
we have discussed so far. You can interrupt a foreground process by typing Ctrl-Z pronounced “control
z” (hold <Ctrl> and hit the z key). Now to put that process back into the foreground, you use the (bash)
command fg.
A background process, by contrast, is one that does not interact with your shell directly. It keeps to itself
and runs silently. To place a process into the background you use the builtin command bg. From the
shell point of view, all processes have a job number. From the system’s point of view, all processes have
a process ID, often called a PID (see the ps command).
To actually be able to catch one of these processes, we need to find a command that lasts longer than an
instant (like date). Lets work with the xterm command which creates an X-terminal window:
<bash>: xterm # Start a fresh xterm

Ctrl-Z # You hit Ctrl-Z and suspended it!
[1]+ Stopped xterm
<bash>: fg # Restarted it with fg

xterm
Ctrl-Z # You suspended it again.. good..
[1]+ Stopped xterm
<bash>: jobs # Find the job number.

[3]+ Stopped xterm
<bash>:
<bash>: bg %3 # Put xterm in the background.

[3] xterm &
<bash>:
You should now have an xterm which behaves normally. The last [3] displayed indicates that its Job
Number is 3. You can use that number to refer to it using fg as in
<bash>: fg %3 # foreground it with fg

Ctrl-Z # Suspend it again.
[3]+ Stopped xterm
<bash>: bg %3 # Put it in the background with bg

[3] xterm &
<bash>: jobs
[1] Running xterm -cr red -ms >&
[2] Running xclock
[3] Running xterm
<bash>: kill -9 %3
<bash>: jobs
[1] Running xterm -cr red -ms >&
[2] Running xclock
[3]- Terminated xterm

9.2. KILLING PROCESSES WITH KILL
The jobs displays all jobs in the shell along with their number and status. The kill -9 %N kills job
number N . The flag (-9) is needed to force a “sure kill” which is sometimes needed. The ampersand
symbol (&) is what you would use to put a command directly into the background with xterm &.
Be careful because some commands want to write errors to the standard output stdout and that could
cause strange results on your screen. To avoid this, you can redirect stdout to /dev/null (or a log file)
AND put it into the background in one gulp:
<bash>: xterm >& /dev/null &

or
<bash>: xterm >& /tmp/xterm-log &
This is advisable if you plan to script your commands, because stdout that has no place to go can stop
your script in its tracks. Experience has shown to always deal with the trash in advance.
9.2 Killing Processes With kill

Now we will display the process numbers with the ps command
<bash>: ps
PID TTY TIME CMD
683 ttyp1 00:00:00 bash
31666 ttyp1 00:00:00 xterm
32716 ttyp1 00:00:00 xclock
<bash>: ps -agl # Get a long list with ps
F UID PID PPID PRI NI VSZ RSS WCHAN STAT TTY TIME COMMAND
000 501 683 672 13 0 2220 804 rt_sig S ttyp1 0:01 [bash]
000 501 31666 683 0 0 3012 1900 wait4 S ttyp1 0:00 xterm -cr r
000 501 32716 683 0 0 2572 1304 do_sel S ttyp1 0:00 xclock
<bash>: top
42 processes: 40 sleeping, 2 running, 0 zombie, 0 stopped
CPU states: 2.3% user, 0.7% system, 0.0% nice, 96.8% idle
Mem: 63484K av, 52328K used, 11156K free, 18020K shrd, 2896K buff
Swap: 220316K av, 4964K used, 215352K free 28804K cached
PID USER PRI NI SIZE RSS SHARE STAT LIB %CPU %MEM TIME COMMAND
664 root 18 0 14824 14M 1540 R 0 1.1 22.7 18:21 X
31797 carinhas 9 0 7060 7060 3276 S 0 0.0 11.1 0:16 xemacs
464 carinhas 0 0 6020 6020 3596 S 0 0.0 9.4 0:03 netscapex
31666 carinhas 0 0 1900 1900 1440 S 0 0.0 2.9 0:00 xterm
32716 carinhas 0 0 1304 1304 1096 S 0 0.0 2.0 0:00 xclock
... <Shift-M> Sorted by memory usage....
Ctrl-C <quit>
Now I have located a renegade process called xclock which seems using way too much memory according
to the top utility. For example, if you want to kill process 32716, do the following:
<bash>: kill -9 32716

[3] Terminated xclock

<bash>:
Study also the man pages for killall, pstree, and proc for the LPI exams.
Next is an abbreviated manual page for kill, but beware that most shells have their own built-in version
of kill.
KILL(1) Linux Programmer’s Manual KILL(1)

NAME
kill - terminate a process
SYNOPSIS
kill [ -s signal ] %job | pid ...
kill -l [ signal ]
OPTIONS
pid .. Specify the list of processes.
-s Specify the signal name or number to send.
-l Print a list of signal names. See signal(7)
Here is the relevant and abbreviated man section of signal(7):

SIGNAL(7) Linux Programmer’s Manual SIGNAL(7)
NAME signal - list of available signals
Signal Value Action Comment

----------------------------------------------------------
SIGHUP 1 A Hangup detected on controlling terminal
SIGINT 2 A Interrupt from keyboard
SIGQUIT 3 A Quit from keyboard
SIGILL 4 A Illegal Instruction
SIGABRT 6 C Abort signal from abort(3)
SIGFPE 8 C Floating point exception
SIGKILL 9 AEF Kill signal
SIGSEGV 11 C Invalid memory reference
SIGPIPE 13 A Broken pipe: write to pipe with no readers
SIGALRM 14 A Timer signal from alarm(2)
SIGTERM 15 A Termination signal
SIGUSR1 30,10,16 A User-defined signal 1
SIGUSR2 31,12,17 A User-defined signal 2
SIGCHLD 20,17,18 B Child stopped or terminated
SIGCONT 19,18,25 Continue if stopped
SIGSTOP 17,19,23 DEF Stop process
SIGTSTP 18,20,24 D Stop typed at tty
SIGTTIN 21,21,26 D tty input for background process
SIGTTOU 22,22,27 D tty output for background process
The letters in the "Action" column have the following meanings:
A Default action is to terminate the process.

B Default action is to ignore the signal.
C Default action is to dump core.

9.3. MANAGING PROCESS PRIORITIES
D Default action is to stop the process.

E Signal cannot be caught.
F Signal cannot be ignored.
G Not a POSIX.1 conformant signal.
9.3 Managing Process Priorities

In Linux and UNIX , all processes have an importance factor called a priority. The priority tells the
system kernel how often to service that process. Obviously some processes are much more critical to
the system than others. Hardware access is more important than email. The smaller the priority value,
the more importance the process is given. Priority values range from -20 to 20, where -20 is the highest
priority and 20 is the least.
You can see the priority of your processes by reading the PRI column in the ps -xl command:
<bash>: ps -xl
000 501 617 612 0 0 2032 504 do_sel S ? 0:03 fvwm
000 501 622 617 0 0 3144 872 do_sel S ? 0:01 xterm -cr r
....etc....
There are two major tools for manipulating process priorities, nice and renice. nice lets you specify
the priority for a command initially, and renice allows you to change the priority after the process is
started. Here’s an exercise for you:
<bash>: man nice

----------------------------------------------------------------
NICE(1) NICE(1)
nice - run a program with modified scheduling priority
SYNOPSIS
nice -n priority command [arguments]
----------------------------------------------------------------
9.3.1 Exercises
<bash>: xclock -rv &
<bash>: nice -10 xclock -rv &
<bash>: /bin/nice -n 10 xclock -rv & # Same as above
<bash>: ps -xl
000 501 617 612 0 0 2032 504 do_sel S ? 0:03 fvwm
000 501 622 617 0 0 3144 872 do_sel S ? 0:01 xterm -cr r
000 501 2855 1321 8 4 2568 1300 do_sel SN ttyp2 0:00 xclock
<bash>: renice 20 2855 # Renice that xclock pid 2855.

2855: old priority 4, new priority 20 # Don’t worry if numbers don’t jive.
<bash>: renice 5 2855 # Please renice to LOWER # ?

renice: 2855: setpriority: Permission denied # Only HIGHER #’s allowed!
<bash>: ps -xl
000 501 617 612 0 0 2032 504 do_sel S ? 0:03 fvwm
000 501 622 617 0 0 3144 872 do_sel S ? 0:01 xterm -cr r
000 501 2937 1321 19 19 2568 1300 do_sel SN ttyp2 0:00 xclock
Note that you can only renice to a higher number (thus lower priority) so the system gives an error
message. Also, the listed PRI takes time to adjust to the NI (nice) value.
LINUX WARNING! tcsh defines its OWN built-in version of nice:
TCSH(1) TCSH(1)
Builtin commands
nice [+priority] [command]
9.4 Cron
Cron is a daemon that can automate tasks by periodically running them based on a time scheme. Its very
useful for maintenance tasks. Cron takes a filename as a script for input that it incorporates internally
and checks periodically. The file is called a crontab file.
9.4.1 Making a Crontab File

The crontab file is by default a–m –h script.
bash style –d It–M –
has extra –E
parameters that tell cron when to start
in mm ohh dda MM oD W COMMAND
e xe
execution. Here is the format: u u r y n e c
te th kd ut
ay ab
l e
Figure 9.1: Crontab File Specification Format
Stars (*) are wildcards which always match, and rely on the other parameters to give cron timing queues.
If you want to execute every day, just put a * in the day column.
A simple example of this would be to run a script above at 12 midnight every night:
# File: crontab-file. Execute me at 00.00 hours EVERYDAY!

0 0 * * * /home/joe user/bin/backup >& /home/joe user/backup-log &
Notice that we have carefully routed stdout to a log file, otherwise cron could hang the job, or send us
all the error by mail.
Another interesting property of the cron time spec is it’s ability to specify numbers in a modulo fashion.
Just divide a * by a number to get cron to execute a periodic command:

9.4. CRON
# File: crontab-file. Execute backups every four hours.

* */4 * * * /home/joe user/bin/backup >> /home/joe user/backup-log 2>&1
9.4.2 Submitting Cron Jobs

Once we have written our crontab file, we can submit it by doing a crontab crontab-file. crontab
-l lists the current crontab entries for you.
<bash>: crontab crontab-file

<bash>: crontab -l
# DO NOT EDIT THIS FILE - edit the master and reinstall.
# (crontabs installed on Mon Feb 7 00:13:36 2000)
# (Cron version -- $Id: crontab.c,v 2.13 1994/01/17 03:20:37 vixie Exp $)
# Execute the job every four hours.
# /home/joe user/crontab-file
* 0 * * * /home/joe user/bin/backup >> /home/joe user/backup-log 2>&1
You may also edit the current crontab file with the crontab -e command, which will throw you into
your default editor. When you exit, the crontab file is automatically updated.
9.4.3 Removing Cron Jobs

You can always remove your cron jobs by using the (-r) flags:
<bash>: crontab -r
<bash>: crontab -l
9.4.4 Exercises
Exercise 1:
Create a simple script named hello that prints out “hello joe user”. It should only have two lines:
#!/bin/bash
echo Hello joe user you slacker...
This file should be placed in your personal ∼/bin directory, and should have the appropriate file mode
for execution (hint: chmod). Make sure to test this script by hand before you go on.
Exercise 2:
Now create a crontab file named crontab-file that will execute hello every 2 minutes. Submit this
crontab file as discussed above and check your mail on the even minutes to see if the scripts work.
Note: Cron’s default behavior is to send any unhandled output to your mail. If mail is not working for
some reason, you may send your output to a file or to your screen.


Chapter 10
Shell Scripting
There’s nothing more aggravating than having to type out long commands over and over again. Shell
scripts are simple text files that list shell commands into a convenient “tool” which you can use more
easily than typing in all those commands over and over.
Now that we know enough shell workings to be dangerous, we need to focus our attention to making
changes permanent. It is essential to have easy to use environments and scripts to do complex tasks,
because otherwise we are stuck repeating numerous small commands daily. The natural starting point is
to customize our .bashrc and .cshrc files with a few aliases and a prompt or two.
10.1 Shell Initialization Files

We’re going to develop simple aliases and shell variables for our bash shell that will make life easier when
you log into your Linux system.
10.1.1 Initialization Files for Bash

Start editing .bashrc now. Write in the following aliases and prompt definitions:
alias m=’less -EMsX ’

alias h=’history | tail -40’
alias d=’ls -l $* | less -EMsX’
alias dt=’ls -blt $* |less -EMsX’
alias Dt=’ls -aglt $* |less -EMsX’
PS1=’[\s]:\u: ’
The $* represents all the remaining variables on the command line except for the calling command. It is
equivalent to all subsequent command line variables, “$1 $2 ...”.
After you modify .bashrc you must do a “source .bashrc” to initialize your changes. Type alias
to see what the system’s definitions, and try out our new aliases.
Remember we can also have a .bash profile which is supposed to carry your environmental variables:
# .bash_profile
# Get the aliases and functions

CHAPTER 10. SHELL SCRIPTING
if [ -f ˜/.bashrc ]; then
. ˜/.bashrc
fi
# User specific environment and startup programs

PATH=$PATH:/usr/local/bin:$HOME/bin
ENV=$HOME/.bashrc
USERNAME=""
export USERNAME ENV PATH
Note the final export at the end of the file.
10.1.2 Initialization Files for Tcsh

Now lets make the analogous aliases in your .cshrc:
alias lester "less -EsX" # Use less as our pager..

alias h ’history \!* | tail -40 | lester’
alias m lester # Used to be more, now more is less.
alias d ’ls -l \!* | lester’
alias dt ’ls -lt \!* | lester’ # Sort by chronological order.
alias Dt ’ls -laFt \!* | lester’ # Same as above, show dot (.xxx) files.
set prompt="%n:%~: " # "user:/location/location/location: "
Try these out in tcsh after executing

<tcsh>: source .cshrc
Now submit these changes by typing source .cshrc. Type alias to see what the system’s definitions,
and try some of them out. Both Tcsh and Csh use .cshrc and .login. The .login file is normally
used to store environment variables and path information, while .cshrc normally hold aliases and prompt
information.
10.2 Utility Shell Scripts

Our first goal is to create a script utility that will print files in a directory called
/home/joe user/address. The syntax will be
phone.shell file1 file2 ....
Because of tcsh’s simple and clear syntax, we will first use the tcsh language to construct a solution.
Afterwards, we show the equivalent bash solution.
10.2.1 Tcsh Solution

The first thing we want to do is to tell the script that it is a csh script, and what the acceptable number
of parameters are. In the following scripts make sure you ALWAYS remove the comment on the first
line, otherwise these examples will not function. In general, students should NOT write any comments
in these examples.
Start by editing a file named ∼/bin/phone.csh:

10.2. UTILITY SHELL SCRIPTS
#! /bin/csh -f # This is not a comment. Remove from 2nd #.

#
set a=$0 # $0 refers to the name of this script.
if($#argv < 1 ) then
echo Usage: $a:t ’Name’
exit(1)
endif
The term set a=$0 refers to the 0th component of the command line, which is the command file,
phone.csh, in our case. The $a:t instructs csh to remove all the leading pathname components,
since we are uninterested in seeing the full path. Test this by invoking the name of the script:
<bash>: phone.csh
Usage: phone.csh Name
Did you remember to chmod our script so that it is executable?

So far the script does nothing but tell us if we have the right format. Next we will add a foreach line
that will give us control for each filename:
foreach name ($*)

echo ~/address/$name
end
Now your script should print each filename you enter on the command-line, as a check in our progress.
Again, test this script and check that the loop works:
<bash>: phone.csh red green blue
/home/joe user/address/red
/home/joe user/address/green
/home/joe user/address/blue
Now that we know it works, replace the echo command with the correct cat phrase. While we are at
it, we will check for the existence of each file with the ( -e filename ) test. The previous foreach clause
should NOW look like:
foreach name ($*)

if (-e ~/address/$name) then # The (-e) flag checks for existence.
cat ~/address/$name
else
echo "no such file $name"
endif
echo "-------------------------" # Separate entries with a line.
end
The final script should resemble the following:
#! /bin/csh -f # This a csh script, omit this comment!

# phone.csh # Comment to give the filename.

set a=$0 # $0 refers to the name of this script.

if($#argv < 1 ) then # If right number of arguements,
echo Usage: $a:t ’name’ # Echo the proper syntax
exit(1) # Exit with error code 1
endif # End if
#
foreach name ($*) # For each name on the command line.
if (-e ~/address/$name) then # If file exists (-e),
cat ~/address/$name # List the file
else # Else
echo "no such file $name" # Barf
endif # End if
echo "---------------------" # Separate entries with a line, cosmetic.
end # Bye-Bye
Lets check our program by creating an /address directory and several files there...
<bash>: mkdir ˜/address

<bash>: cd ˜/address
<bash>: echo hello1 says 123 > hello1 # Creates a file called hello1.
<bash>: phone.csh hello1 hello2 hello3
hello1 says 123

------------------------
hello2 says 456
------------------------
hello3 says 789
------------------------
10.2.2 Bash Solution

The bash solution to the exercise is not as obvious as the csh one because of bash’s bizarre syntax.
Call this file phone.bash:
#! /bin/bash # This is a bash script. Omit comment.

if [ "$#" -lt 1 ] ; then # If number of arguments too small
echo Usage: $0 args # State the correct usage
exit 1 # Exit with error status
fi # Note "fi" closes "if" statements
for i in $@ ; do # Start a "do" loop on the $@ array
if [ -f ~/address/$i ];then # If the file exists.
cat ~/address/$i ; # print it,
else # Else
echo There is no file $i # Error message if absent.
fi # End if
echo -------------- # Print a separator between files
done # "done" closes the "do" loop

10.3. GENERAL SCRIPTING
10.2.3 Exercises
Exercise 1: Make the phone.bash version above work.
Exercise 2: Add an extra error message line if the file does not exist.
10.3 General Scripting

In this section we will create a several script programs named “doubletake” that will remove repeat entries
in a list. We will use several of our newly learned languages to accomplish that task.
10.3.1 awk version

Edit doubletake and put in the following lines:
#! /bin/csh -f
# This script remove double in a sorted file. See the uniq manual.
set file1 = $1 # $1 is going to be the filename.
cat $file1 | awk ’{ if ($0 != last ) {print $0; }; }{ last = $0; }’
The script first stores the first command argument in $1’. It then cat’s the file into awk. Notice you
need the initial line of
#! /bin/csh -f
to tell the system that this is indeed a csh script.
Now the script that awk uses does 2 things. It first checks to see if the variable last was seen, if not,
prints it. The second thing it does is to store the current line in last. Of course you could just type the
whole second line into the command line, but that’s not much fun if you have to do it 9 times a day.
10.3.2 Perl Version

#! /usr/bin/perl -w
my $line = ""; # declare this variable.
foreach $filename (@ARGV) # cycle through each filename in the ARGV list.
{
open (SF, $filename) or die "Can’t open $filename" ; # open the file and complain if not.
while (<SF>) # For every line in the file
{
chomp($_); # Remove tailing whitespace and/or linefeeds.
if ( $line ne $_ ) # ne = not equal
{
print "$line \n"; # Yes, print me.
}
$line = $_ ; # Reset $line.
}
}

10.4 Script Automation

Shell scripts are great tools when you can use them, and can save a lot of time. If you need to run
applications or backups at odd hours or nightly, you really want to automate the task by using the cron
tools, which we have already covered in chapter 9.4.
The following exercise describes how to set up a script program to do your dirty work while you are
playing golf or sleeping.
10.4.1 Getting Scripts in Order

Before you submit a job for cron to handle, you must make sure you can run the job by hand. This
involves making every command self sufficient by providing complete paths to the commands and setting
the right permissions. Lets take backups for example. You need to have privileges to write to the tape
device as a normal user unless you plan to run your jobs as root .
Here is a typical script, /home/guest/bin/backup.sh that would back up your home directory to the
floppy device /dev/fd0. In this example, the floppy is being used as a character device just like a tape
drive would be used:
#! /bin/bash
# This file is named backup.sh
cd /home/guest # Make sure you are at home.
/bin/tar cvfz /mnt/fd0 . # Back up your home directory
echo All done boss!
In this example, we have specified all paths to commands. This is required since your normal path/shell
is not invoked in a script like this one. This script will create a lot of output which must be sent into a
log file or /dev/null later when we make our crontab file.
10.4.2 Testing the Scripts

You must make sure you can run the scripts by hand before you try to invoke cron. This is typical for
all programs that must be automated. Also make sure as mentioned before that all commands are given
by their full path names. BTW, why doesn’t the echo command need a path?
10.4.3 Exercises
Ex 1: Once you are confident that your script will run on its own two feet, you are ready to set it up with
cron. Create a crontab file /home/guest/crontab.backup and submit it with
<bash>: crontab crontab.backup
as done in section 9.4. Ask for help if you get stuck.

Chapter 11
Networking and The Internet
Linux comes with all the TCP/IP networking goodies you would expect from a full-featured
UNIX box. It has telnet, ftp, mail, and many other services. Here’s a quick list of the
bread-n-butter commands:
Networking Terms
TCP/IP Transmission Control Protocol / Internet Protocol
udp Best effort packet delivery protocol
tcp Reliable byte stream protocol
rpc Remote procedure call
ARP Address Resolution Protocol
route a network route to another machine
host Any machine on the network
client The host that wants data or a service
server The host that provides data or a service
Remote access and file transfers

telnet RemoteHost Remote login, often disabled for security reasons
rlogin RemoteHost Login to RemoteHost (insecure)
rsh RemoteHostcommand Execute command on remote host. Used in conjunc-
tion with .rhosts file. Often disabled for security rea-
sons.
ftp RemoteHost File Transfer Protocol transfers files between hosts
sftp RemoteHost Secure file transfer protocol
rcp fromfile tofile Remote copy (insecure)
scp file RemoteHost Secure copy file from here to RemoteHost
scp Host1:f1 Host2:f2 Secure copy f1 from Host1 to Host2
Email
.forward File in user’s home directory containing mail forward-
ing address
mail user[@RemoteHost] [¡ file ] Mail user agent
pine Mail user agent
elm Mail user agent
mutt Mail user agent

CHAPTER 11. NETWORKING AND THE INTERNET
System Information:
hostname -f Set or show the system’s host name
hostid Report the numeric identifier of the local host
uname -a Report system information
uptime Report time since last reboot
du -h Estimate file space usage at and below the cwd
df -h Report file system disk space usage
ps aux Report process status
top Report top processes
Superuser/Root:
su [user] Switch to different user (user must be in /etc/passwd
file)
sudo command Execute command as root (User must be in
/etc/sudoers file)
chown owner : group file Change file/directory ownership
halt This and the next two commands reboot the system
(must be superuser)
reboot
shutdown
Ctrl-Alt-Delete Reboot the sytem
TCP/IP Diagnostics:
Some of these commands may require Superuser priviledges.
/etc/hosts Static table lookup for host names.
/etc/resolv.conf Static table with name of the DNS
/etc/services Internet network services list.
ifconfig Display or configure local TCP/IP network interface con-
figuration
ping RemoteHost Send a “ping” to a remote host to check connectivity
traceroute RemoteHost Print the route packets take to network host
ipmaddr
iptunnel
arp Display or modify IP-to-Physical address translation ta-
bles used by ARP
rarp
route Print out the local routing table (check /sbin/route)
dig RemoteHost Send domain name query to DNS
nslookup RemoteHost Send domain name query to DNS
tcpdump -i ethX Listen to network traffic on ethX
ethereal Capture and analyze network traffic
netstat Display protocol statistics
netwatch Monitor network connections
nmap Network exploration and security monitoring
whois Client for the whois directory service
FTP examples
• ftp hostname
Name userName
Password userPassword

11.1. NETWORK COMMUNICATION
11.1 Network Communication

This section covers network communication from a user point of view. We won’t talk much about how
to set up hardware or server configurations. This has already been done by your administrator.
11.1.1 Checking Network Status

Sometimes it is unclear if your network is working correctly. There are some easy commands that you
can use to check to see if a computer is “alive” on the network. The first and most important command
is ping which will send a test message to the computer who should respond:
[LocalHost]/home/joe user:ping groucho

PING groucho (192.168.1.3): 56 data bytes
64 bytes from 192.168.1.3: icmp_seq=0 ttl=255 time=0.7 ms
....
This shows that the host groucho is alive and well. It provides us with some timing information which
indicates the speed of the connection between LocalHost and RemoteHost.
If your DNS (see next section) is not working, you may not be able to ping by name, but perhaps by
number. You can also ping by IP number like this:
[LocalHost]/home/joe user:ping 192.168.1.3

PING 192.168.1.3 (192.168.1.3): 56 data bytes
...etc...
If you don’t get any response, or do get an error, something is wrong with the network and you will not
be able to communicate with that computer.
11.1.2 DNS, Dig, and Nslookup

The Domain Name Service also known as DNS allows you, the users, to translate names like www.yahoo.com
into a number like 216.32.74.52 which is needed for your computer to communicate over the network.
Networks are controlled by gremlins that only use numbers not names. Thus DNS is very important if
you are going to use the network at all.
Who provides DNS service? Most of the time your ISP provides this service, especially if you are using a
modem, cable modem, DSL, etc. If you are in a large corporate setting, your own company network will
provide DNS.
The main tool you need to check that your DNS is working is nslookup which can translate names to
numbers or vice-versa:
[LocalHost]/home/joe user:nslookup www.io.com

Server: flure.pair.com
Address: 209.68.1.159

Name: www.io.com
Addresses: 199.170.88.21, 199.170.88.41, 199.170.88.39
Nslookup tells us there are 3 names that go with www.io.com. Large sites like www.io.com and
www.yahoo.com often have many numbers because they maintain several servers to handle all the re-
quests to their very busy sites. The server flure.pair.com is our DNS server as seen by our local DNS
setup.
DNS can also translate numbers into names:
[LocalHost]/home/joe user:nslookup 199.170.88.21

Address: 209.68.1.159
Name: www-02.io.com
Address: 199.170.88.21
This was one of the www.io.com sites listed in the first example. If DNS can’t find your name, there is
little chance you can connect through the internet to it:
[LocalHost]/home/joe user:nslookup belse.com

Address: 209.68.1.159
*** flure.pair.com can’t find belse.com: Non-existent host
Sorry, you are SOL. No can do.. (unless you know the number already) Maybe try www.belse.com.
Sometimes you can connect by IP number even when DNS doesn’t work. Try to connect to www.io.com
with one of the numbers we found. Note that Linux has a text browser called Lynx that is many-times
convenient.
11.1.3 The “R” Commands

The so-called R-Commands are rlogin, rcp, rsh, and friends. These are powerful and dangerous
commands, and should be used with great care. Many systems exposed to the Internet disable, as they
should, these commands because data is completely open to public scrutiny over that network. Now day,
some of these commands have secure versions whose data is encrypted, or the commands are replaced by
ssh, the secure shell.
In order to execute remote commands on a remote machine without passwords, you set up a file called
.rhosts (in your home dir) that looks like the following:
# .rhosts file for RemoteHost

LocalHost joe user
and on the local machine a complimentary setup allows you to work from there.

11.2. USING X-WINDOWS OVER THE NETWORK
# .rhosts file for LocalHost

RemoteHost joe user
Now try to copy a file from here to there:
rcp filename RemoteHost:

rcp filename RemoteHost:. # Same as above.
Now copy a file from RemoteHost to LocalHost:
rcp RemoteHost:fileA . # Copy the file to . with same name.

rcp RemoteHost:fileA fileB # Renames the file
Now user rlogin to log into RemoteHost:
rlogin RemoteHost
Last login: Sat Jan 15 02:53:13 from beppo
You have mail.
[RemoteHost]/home/joe user:ls -agl
total 3236
drwxr-xr-x 20 joe user users 4096 Jan 14 09:45 .
drwxr-xr-x 5 root root 4096 Jan 14 08:42 ..
-rw------- 1 joe user users 1748 Jan 14 06:59 .ICEauthority
-rw------- 1 joe user users 101 Jan 14 08:35 .Xauthority
-rw-r--r-- 1 joe user users 1506 Jan 14 08:18 .Xdefaults
-rw------- 1 joe user users 133 Jan 11 15:56 .bash_history
....etc....
11.2 Using X-Windows Over the Network

One great thing about X is that it has networking built right into it. You can use any X application from
anywhere on your network. All you need to do is set a few parameters that allow communications.
11.2.1 Connecting X to another client

There are two things you must do to get X to work over a network. First you need to allow other machines
to connect to your workstation. One way to do this is to use xhost RemoteHost, where RemoteHost
is the machine you want to communicate with.
Second, you must telnet (or ssh) to a another workstation, and set the DISPLAY environment variable
to your workstation like this:
[LocalHost]/home/joe user:xhost + RemoteHost

[LocalHost]/home/joe user:telnet RemoteHost
Red Hat Linux release 6.1
Kernel 2.2.12-20 on an i586
login: joe user
Password: ******

Last login: Thu Jan 13 12:13:37 from bistro.rr.com

[RemoteHost]/home/joe user:setenv DISPLAY LocalHost:0
[LocalHost]/home/joe user:xcalc -title RemoteHost
Ctrl-C
[RemoteHost]/home/joe user:exit
[LocalHost]/home/joe user:
Remember that RemoteHost and LocalHost need to be changed to whatever your workstation is named.
What you should have now is a process that is run by the remote system! It should be clear that this
is powerful. In the next chapter we’ll talk about other networking features.. If you put a line like the
following in your .login or .bashrc, your DISPLAY variable will be set automatically for you. Play with
this for a second:
<tcsh>: setenv DISPLAY ‘who am i | perl -ne ’/$(.+)$/ ; print $1’‘”:0”

or in bash,
<bash>: set DISPLAY= ‘who am i | perl -ne ’/$(.+)$/ ; print $1’‘”:0”
11.2.2 Securing Your X Connection

Above we talked about xhost and how it allows other machines to connect to your X-Server on your
workstation. There is a security flaw that allows ALL people from the RemoteHost to connect to your
workstation. This “hole” has been used to exploit many a computer.
Fortunately there is a secure alternative to using xhost called xauth. The way to use xauth is to first
execute xauth list on the RemoteHost, grab the information it produces with your mouse (left click
and drag), and then paste it (middle click) on your workstation right after typing xauth add.
[RemoteHost]/home/joe user:xauth list

bistro.marx:0 MIT-MAGIC-COOKIE-1 6c7d8ba30a
bistro/unix:0 MIT-MAGIC-COOKIE-1 7bf452301a230498
localhost:0 MIT-MAGIC-COOKIE-1 a43f0989db
localhost:10 MIT-MAGIC-COOKIE-1 3c08b9241d
On your (local) workstation:
<bash>: xauth add bistro/unix:0 MIT-MAGIC-COOKIE-1 7bf452301a230498
11.3 Network Security for Users

Network security from a user point of view is simple: Don’t use easy-to-guess passwords, and don’t leave
the “R” commands open to attack.
Hard to guess passwords don’t resemble any word and should have non alphabetic characters in them.
Hello is not a good password, but h3l9Xz is. Hackers always use crack which uses a dictionary to guess
passwords, and Hello would be cracked in a few seconds.
As far as security is concerned, you need only remember a couple of things.

11.3. NETWORK SECURITY FOR USERS
• Always keep your home directory (˜) protected.

<bash>: chmod -R 700 ∼/
• Don’t use R-commands over an insecure network, as hackers can see most everything. If you
have secure shell OpenSSH and secure versions of R-commands, you are better off. Check http:
//www.openssh.org for info on OpenSSH.
• Programs like telnet show your passwords over the Internet. Use OpenSSH from http://www.
openssh.org
• Never tell anyone your password, and if you do, change it the next day.
• chmod 600 .rhosts

• Change your password every few months.
• Never write your password down or identify it as a password.
• Pick an account names and passwords that are difficult to guess. Passwords should have uppercase,
lowercase, and numeral characters. Do not use a dictionary word as a password or even part of a
password. You can check your password tables agains crack at ftp://ftp.cerias.purdue.edu/
pub/tools/unix/pwdutils/crack.
• Check permissions on files periodically. Most files should be 600 or 700.


Chapter 12
Program Development in
Unix/Linux
Program Development Tools

gas GNU Assembler
cc file.c Compile a C program
gcc file.c Compile a C program
g++ file.cpp Compile a C++ program
gdb The GNU program debugger of C and C++
make Maintain, update, and regenerate program and files based
on user defined dependencies found in the file makefile
cat -A makefile Show non-printing characters (especially tabs and tailing
spaces) in makefile
rcs options file ... The revision control system RCS
ci Check a file into the RCS directory
co Check out a file from the RCS directory
cvs [opts] cmd [opts] [args] The concurrent version system (CVS) which allows you
to keep old versions of files, a log of who, when, and why
changes occurred, etc., like RCS and SCCS
script Capture terminal interactions into a file called typescript
exit Exit a script session
Many compilers are available for Linux including versions of C++, C, Perl, Python, awk, and many
others. It also has text formatting solutions of TEX and LATEX which is what we use to format this
document. There are many other freely available additions like Java, Fortran, and Lisp.
Here is what our systems spits out with man -k compiler:
cccp, cpp (1) - The GNU C-Compatible Compiler Preprocessor.

f4rpcgen (1) - an RPC protocol compiler
g++ (1) - GNU project C++ Compiler
gcc, g++ (1) - GNU project C and C++ Compiler (egcs - 1.1.2)
gcc, g++ (1) - GNU project C and C++ Compiler (egcs-1.1.2)
tic (1m) - the terminfo entry-description compiler
zic (8) - time zone compiler
B (3) - The Perl Compiler
B::Bytecode (3) - Perl compiler’s bytecode backend

CHAPTER 12. PROGRAM DEVELOPMENT IN UNIX/LINUX
B::C (3) - Perl compiler’s C backend

B::CC (3) - Perl compiler’s optimized C translation backend
B::Deparse (3) - Perl compiler backend to produce perl code
O (3) - Generic interface to Perl Compiler backends
Here is what our systems gives for man -k language:
Tcl (n) - Summary of Tcl language syntax.

bc (1) - An arbitrary precision calculator language
gawk (1) - pattern scanning and processing language
perl (1) - Practical Extraction and Report Language
perlxs (1) - XS language reference manual
python (1) - an interpreted, object-oriented programming language
12.1 C & C++ Compilers

12.1.1 C programs
Edit program try.c with favorite text editor.
#include <stdio.h>
int main() {
printf("running my first program\n");
return(0);
}
Compile program
<bash>: cc try.c
Run program
<bash>: a.out
Print program
<bash>: lpr try.c
Capture program and output
<bash>: script
<bash>: cat try.c
<bash>: cc try.c
<bash>: a.out
<bash>: exit
Print session
<bash>: lpr typescript
12.1.2 C++ programs

Edit source program in favorite editor then

12.2. DEBUGGING WITH GDB
<bash>: gpp YourSourceFile.cpp

<bash>: ls
a.out YourSourceFile
<bash>: a.out
Your program in action
12.1.3 Command-Line Arguments
12.1.4 Standard C Libraries
12.1.5 Libraries and Archives
12.1.6 Error Handling in C
12.1.7 Error Recovery
12.2 Debugging with gdb

• Interactive debugging
• Basic dbx commands
– trace
12.2.1 Examining Core Dumps

Sample C session
<bash>: cc
Sample C++ session
<bash>: g++
12.2.2 The Debugger

Sample debugging session
<bash>: cc -
<bash>: gdb
12.3 The Unix make Facility

make is a command generator which uses a description file to create a sequence of commands for execution
by the Unix shell. It is commonly used as a tool for maintaining component modules in a software system
and for building an executable from group of files.
The make command performs actions defined in a description file usually named makefile or Makefile.
The makefile is placed in the root directory of the collection of files comprising the application. The make
utility determines which files to update by using the list of dependences described in the makefile and a

file’s time stamp field (last-modified time). The makefile consists of suffix definitions, macro definitions,
dependencies, and commands. Each command behaves as if it were invoked within a shell.
Entries in a makefile have the form:
target : prerequisites
(tab) commands
where both the prerequisites and commands are optional. The commands are executed only if the target
is not upto data with respect to the prerequisites. The target is dependent on one or more files called
prerequisites each of which may have other files as prerequisites. The description file defines all of the
intermediate steps necessary to accomplish the task. The first character of a command line must be a
tab and no other line may begin with a tab. The following command may be used to check the tabs in
the description file.
<bash>: cat -v -t -e makefile
or simply
<bash>: cat -A makefile
The makefile is run by entering the command
<bash>: make
or
<bash>: make target
The make utility determines what parts of a program need to be rebuilt and then carries out the appro-
priate set of actions producing the latest version of the target.
For more details than are provided in this section, see [4] or visit the GNU Project web site server:
www.gnu.org/software/make/make.html.
The example in Subsection 12.3.3 is a portion of the makefile used in the preparation of this text.
12.3.1 The Description File

The makefile consists of one or more lines each of which describes either a dependency rule, a command,
a macro, or a suffix rule. A dependency is a target followed by a list of items on which the target is
dependent. A command is an action that following a dependency describes how the target file is created
from the dependencies. The line containing a command always begins with a tab character. The target
files are intermediate files and the ultimate executable file that results from linking the object files
together. The dependent files are source files ( .c, .cpp or .h files) or object files (.o). A macro is a name
for a set of files. The syntax of the makefile rules is described in Table 12.1.
Commands
Commands have the form:
TAB command [; command]
Additional commands may be placed on separate lines or separated with semicolons if placed on the same
line.

12.3. THE UNIX MAKE FACILITY
makefile ::= suffix rule | dependency rule | command | macro definition

suffix rule ::= .suffix[.suffix] :
dependency rule ::= target : [prerequisites] [; [command]]
command ::= tab Shell command
macro definition ::= name = string
Table 12.1: Makefile syntax
Dependence Rules
Dependence rules have the form:
target : prerequisites
(tab) command
The prerequisites may be placed on separate lines provided the previous line ends in a backslash. The
list of prerequisites may be followed by a command which is either preceded by a semicolon or entered
on a new line.
Macros
A macro definition has the form:
NAME = string
It is customary to use all caps in the name of the macro. A tab may not precede a macro definition and it
is terminated by a new line. A macro must be defined before is it used and it is referenced by ${NAME}.
Suffix Rules
Suffix rules have the form:
.suffix.suffix :
(tab) command
Files ending with the first suffix can be prerequisites of files that have the same name but end with the
second suffix. No spaces are allowed between the suffixes. The command used to transform a file with
the first suffix to one of the second suffix may be placed on the next line. See the example in subsection
12.3.3.
The suffix dependences for C and C++ are built into make so need not be used.
12.3.2 Short Example
12.3.3 Larger Example

The following example is the make file used in preparing this text and it illustrates these concepts.

# Macro definitions
SHELL = /bin/sh # Limit commands to Bourne shell
# Suffixes significant for typeset documents

.SUFFIXES : .aux .dvi .idx .log .tex .toc .pdf .ps .tgz
# Suffix relationships and build commands

# Create .dvi files
.tex.dvi: #latex must be run twice to resolve forward references
latex $< ; latex $<
# Create .ps (postscript) files from the .dvi file

.dvi.ps :
dvips $<
# Create .pdf files from the .tex file
.tex.pdf :
pdflatex $<
# Subdirectories
COMP = Compiler
DOC = DocPrep
FDIR = Fundamentals
PROG = Programming
# File dependencies (correspond to target input and includes)

# Touch command necessary to force rebuild
# Lines deleted that refer to files use in other texts
# Fundamentals
fundamentals.tex : definitions.tex defsAaby.tex\
${FDIR}/X-Windows.tex\
${FDIR}/bibliography.tex\
${FDIR}/commandline.tex\
${FDIR}/copyright.tex\
${FDIR}/environment.tex\
${FDIR}/filesystem-management.tex\
${FDIR}/filesystem.tex\
${FDIR}/glossary.tex\
${FDIR}/introduction.tex\
${FDIR}/process.tex\
${FDIR}/scripting.tex\
${FDIR}/security.tex\
${FDIR}/software.tex\
${FDIR}/tcpip.tex\
${FDIR}/terms.tex\
${FDIR}/text-tools.tex\
${FDIR}/unixCommands.tex
touch fundamentals.tex
# Build all of the documents.

12.4. CONCURRENT VERSION CONTROL (CVS)
# This make file serves to build four texts
all : compiler docprep fundamentals programming
# Build individual documents
compiler : compiler.tex compiler.dvi compiler.ps compiler.pdf

docprep : docprep.tex docprep.dvi docprep.ps docprep.pdf
fundamentals : fundamentals.tex fundamentals.dvi fundamentals.ps fundamentals.pdf
programming : programming.tex programming.dvi programming.ps programming.pdf
# Create src package - run make clean first

release :
cd ..; /bin/tar cfz linux.tgz Linux; mv linux.tgz Linux/.
# Remove all non-source files.

clean :
/bin/rm -f *.aux *.dvi *.idx *.log *.toc *.pdf *.ps *.tgz
12.4 Concurrent Version Control (CVS)

CVS (Concurrent Versions System) is a tool that is used to track changes to any thing that can be
stored as a set of files. It permits multiple developers to coordinate changes, track versions, and permit
simultaneous development on different versions of the same code.
For more details than presented here, see [5] or visit www.cvshome.org.
Concurrent Version Control (CVS)

cvs The Concurrent Versioning System (CVS) pro-
vides version
control in a multi-developer, multi-directory environ-
ment.
cvs [*] -H [cmd] CVS help option
CVSAREA= sanbox .profile or .bashrc environmental variable.
CVSROOT= repository for the sandbox and source repository respectively
<tr>
<td colspan="2">CVSAREA={\it sandbox} 
CVSROOT={\it sourceRepository}</td>
<td>.profile or .bashrc file environmental variable entries for sandbox
and source repository respectively.</td>
</tr>

<tr>
<td colspan="2">setenv CVSAREA {\it sandbox} 
setenv CVSROOT {\it sourceRepository}</td>
<td>.cshrc file environmental entry for sandbox and source respository
respectively.</td>
</tr>
<tr>
<td>cvs</td>
<td>[-d {\it sourceRepository}] init</td>
<td>Create the source repository</td>
</tr>
<tr>
<td>cd 
cvs</td>
<td>$CVSAREA/{\it projectDirectory} 
import -m {\it msg repository vendor release}</td>
<td>Check project files into repository from the sandbox.</td>
</tr>
<tr>
<td>cd 
cvs</td>
<td>$CVSAREA/{\it projectDirectory} 
checkout {\it projectDirectory}</td>
<td>Check project files out of the repository and into the sandbox.</td>
</tr>
<tr>
<td>cvs</td>
<td>commit -m msg file</td>
<td></td>
</tr>
12.5 Packaging Your Software - Tar

The file compression utilities are used to decrease storage requirements and to reduce the time it takes
to transmit files.
compress file compressedFile.Z The compress utility. Compressed files should have a .Z suffix.
uncompress file.Z The uncompress utility.
gzip file zippedFile.gz The GNU zip utility. Zipped files should have a .gz suffix
gunzip file.gz The GNU unzip utility.
The tar (tape archive) utility is used to construct a linear representation of the unix hierarchical file
structure and restore the hierarchical structure from the linear representation. In addition, it may be
used to compress and decompress the files. It was originally designed to save file systems on tapes as a
backup.
tar cf theArchive.tar theDirectory - creates a tar file of theDirectory and its contents an
places it in the working directory.

12.6. SESSION CAPTURE
tar cfz theArchive.tgz theDirectory - creates a gziped tar file of theDirectory and its contents
an places it in the working directory.
tar xf theTarFile.tar - extracts the tar file into the working directory, duplicating the tarred
directory structure.
tar xfz theTarFile.tgz - gunzips and extracts the tar file into the working directory
tar tzf theTarFile.tgz - List of the contexts of the tar file
tar xf theTarFile FileOfInterest - extract the file of interest from the tar file.
12.6 Session Capture

The following illustrates how to capture and print a terminal session to provides a program listing and
evidence of compilation and execution.
<bash>: script
<bash>: cat yourProgram.c
...
<bash>: cc yourProgram.c
<bash>: a.out
...
<bash>: exit
<bash>: lpr typescript

Consult the following websites for current software information:

http://www.freshmeat.net for Linux software.
http://www.sourceforge.net for OpenSource software.
http://www.newsforge.com Open Source News.
http://www.seul.org Linux software for education and science.
http://www.slashdot.org Don’t forget Slashdot.
12.7 Exercises
1. Create makefile for a small C or C++ program. The program should consist of several files. Your
makefile should insure that only the necessary files are recompiled after changes.
2. Package the program and makefile of the first exercise in a gzipped tar file.

Appendix A
Glossary
alias . . . . . . . . . . . . . . . . . . . . A shortcut name for another command

ASCII . . . . . . . . . . . . . . . . . . American Standard Code for Information Interchange = text
background . . . . . . . . . . . . Processes work silently here and dont bother you
bash . . . . . . . . . . . . . . . . . . . . The GNU Bourne-Again SHell.
bin . . . . . . . . . . . . . . . . . . . . . Refers to an executable ”binary”, usually a directory
cwd . . . . . . . . . . . . . . . . . . . . . Your Current Working Directory
cursor . . . . . . . . . . . . . . . . . . The active point for editing, usually a highlighted rectangle
daemon . . . . . . . . . . . . . . . . A program in the background and provides a service, like HTTPD
device . . . . . . . . . . . . . . . . . . A physical or logical component of a computer
directory . . . . . . . . . . . . . . . A folder in the computer
DNS . . . . . . . . . . . . . . . . . . . . Domain Name Service: domain names for IP numbers & vice-versa
environment . . . . . . . . . . . A set of parameters your shell works in
execute . . . . . . . . . . . . . . . . To enter a command
ext2 . . . . . . . . . . . . . . . . . . . . The native Linux filesystem
file permissions . . . . . . . They control who reads, writes, and executes files
foreground . . . . . . . . . . . . . Where processes with your console
FTP . . . . . . . . . . . . . . . . . . . . File Transfer Protocol, used to move files over networks
GID . . . . . . . . . . . . . . . . . . . . Group ID of a file or process
GUI . . . . . . . . . . . . . . . . . . . . Graphical User Interface. Think Windows, but better!
$home . . . . . . . . . . . . . . . . . . Your home directory
host . . . . . . . . . . . . . . . . . . . . A computer on a network
job . . . . . . . . . . . . . . . . . . . . . A process running in your shell
kernel . . . . . . . . . . . . . . . . . . The core software that runs Linux
man . . . . . . . . . . . . . . . . . . . . The MANual pages
mount . . . . . . . . . . . . . . . . . . Attach a disk partition to Linux
mount point . . . . . . . . . . . An empty directory you attach (mount) a partition to
network . . . . . . . . . . . . . . . . A series of communication channels for computers
partition . . . . . . . . . . . . . . . A subdivision of a hard disk. A slice of pie.
path . . . . . . . . . . . . . . . . . . . . /This/is/where/you/are/ in the filesystem (think folder)
PID . . . . . . . . . . . . . . . . . . . . Process ID
priority . . . . . . . . . . . . . . . . Number indicates how important a process is to the kernel
process id . . . . . . . . . . . . . . Every running program has such a number
prompt . . . . . . . . . . . . . . . . . That piece of text just before you type commands. Customizable.
regular expression . . . . A pattern used for searching text.
route . . . . . . . . . . . . . . . . . . . The path data takes through a nework.

APPENDIX A. GLOSSARY
script . . . . . . . . . . . . . . . . . . . An shell program much like a ”batch file” in MS-DOS.

shell . . . . . . . . . . . . . . . . . . . . A program that reads/executes command from users.
server . . . . . . . . . . . . . . . . . . A computer that provides data and/or services
stderr . . . . . . . . . . . . . . . . . . Standard Error
stdin . . . . . . . . . . . . . . . . . . . Standard Input
stdout . . . . . . . . . . . . . . . . . Standard Output
string . . . . . . . . . . . . . . . . . . Just a text word
swap . . . . . . . . . . . . . . . . . . . Memory on a disk
tcsh . . . . . . . . . . . . . . . . . . . . The C-Shell, very friendly
UID . . . . . . . . . . . . . . . . . . . . User ID. Each user has one and files get tagged with it.
username . . . . . . . . . . . . . . Your login name
user . . . . . . . . . . . . . . . . . . . . A computer user
xterm . . . . . . . . . . . . . . . . . . A console program for X-Windows

Appendix B
Unix/Linux Commands by Location
Commands are found in several different locations as specified by the filesystem hierachy standards FSH
2.2; see www.pathname.com/fhs. Not all commands in the standard are listed here. Not all commands
listed here are covered in the text.
Location of Commands in the Filesystem Hierarchy

/bin Essential user commands (for use by all users)
/sbin Essential system commands for use by system administra-
tor
/usr/bin Primary directory of user commands
/usr/sbin Non-essential standard system commands used by system
administrator
/usr/local/bin Host specific commands (for use by all users)
/usr/local/sbin Host specific system commands used by system adminis-
trator
/home/userid/bin User’s frequently used binaries and scripts.
The following are an abbreviated listing of popularly used commands found in the directories.
/bin: Commonly used essential commands

arch Print machine architecture
awk Pattern scanning and processing language
bash GNU Bourne-Again Shell
cat Concatenate files and print on the standard output
chgrp Change group ownership of files
chmod Change access permissions (mode) of files
chown Change ownership of files
cp Copy files and directories
date Print or set system date and time
dd Convert and copy a file
df Report filesystem disk space usage
dir List directory contents (also ls and vdir)
du Estimate file space usage
echo Display line of text
egrep Print lines matching a pattern (grep -E)
fgrep Print lines matching a pattern (grep -F)
grep Print lines matching a pattern

APPENDIX B. UNIX/LINUX COMMANDS BY LOCATION
groups Print the groups a user is in

gunzip compress or expand files
gzcat compress or expand files
gzip compress or expand files
hostname Set or print the name of the current host
kill Terminate a process
last Show listing of last logged in users
ln Make links between files
ls List directory contents (also dir and vdir)
mail Send and receive mail
mesg Write control access to terminal
mkdir Make directories
more file View file contents a page at a time
mv A B Move (rename) file A to B
netstat Print network informaiton and statistics
ping Send ICMP ECHO REQUEST to network hosts
pstree Display a tree of processes
ps Report process status
pwd Print name of current/working directory
rmdir Remove empty directories
rm Remove files or directories
sed Stream editor
sh Execute following script in the bash shell
su Change user Id or become super-user
sync Flush filesystem buffers
tar Tape archiving utility
tcsh C shell with file name completion and command line edit-
ing
uname -a Print system information
wall Send a message to everybody’s terminal
wc Print the number of bytes, words, and lines in files
zcat Compress or expand files
zmore File rerusal filter for viewing compressed text
/sbin: Commonly used essential system commands

arp Maninpulate system ARP table
arping Send ARP REQUEST to a neighbor host
halt Stop the system (also reboot, poweroff
ifconfig Configure a network interface
init Process control initialization (also telinit)
rarp Manipulate system RARP table
reboot Stop the system (also halt and poweroff
route Show/manipulate the IP routing table
runlevel Find the current and previous system runlevel
shutdown Bring the system down
telinit Process control initialization (also init)
/usr/bin: Commonly used commands

Commands available depend on the distribution.
aspell Spell checker (also ispell)

autoconf Generate configuration scripts
automake Creates Makefiles from template files
awk Pattern scanning and processing language (also gawk)
bison GNU parser generator (yacc replacement)
c++ C and C++ compiler (also cc, gcc and g++)
cancel Cancel print jobs
cat file(s) Concatinate files, result to standard output
cc C and C++ compiler (also c++, gcc, and g++)
chfn Change user name and information
chgrp Change group ownership of files
chown Change ownership of files
gunzip compress or expand files
compress compress files
cp Copy files and directories
crontab Maintain crontab files for individual users
cvs Concurrent versions system
dc An arbitrary precision calculator
dd Convert and copy a file
df Report free space remaining in the file system
diff Find differences between two files
dig DNS lookup utility
dir List directory contents (also ls and vdir)
du Report disk space used in the current directory and all
subdirectories
emacs Emacs text editor and system interface
ethereal Interactively browse network traffic
find Search for files in a directory hierarchy
flex Fast lexical analyzer generator
fortune Print a random adage
g++ C and C++ compiler (also c++, cc, and gcc)
gawk Pattern scanning and processing language (also awk)
gcc C and C++ compiler (also c++, cc, and g++)
gdb The GNU debugger
ggv Display postscript and pdf files
hostid Print the numeric identifier for the current host
hostname Set or print the name of the current host
host DNS lookup utility
info Read Info documents
kill Terminate a process
lastlog Examine lastlog file
less Similar to more but allows backward movement
lex Lexical analyser generator (see Flex)
ln Make links between files
locate Index and search for files on your system
logname Print user’s login name
look Display lines beginning with a given string
lpq Show printer queue status
lprm Cancel print jobs
lpr Print file using default printer
lpstat Print printing system status information

lp Print files
ls List directory contents (also dir and vdir)
lynx Text terminal browser for the WWW
m4 Unix macro processor
make Utility to maintain groups of programs
man Format and disply the on-line manual pages
mkdir Make directories
mv Move (rename) file
nasm Portable 80x86 assembler
nice Change execution level of task
nohup Run a command immune to hangups, with output to a
non-tty
nslookup Deprecated. Use dig instead
perl Practical extraction and report language
pg Browse pagewise through text files
pr Convert text files for printing
pstree Display a tree of processes
python Interpreted, interactive, O-O programming language
rcs Revision control system
renice Alter priority of running processes
rmdir Remove empty directories
rm Remove files or directories
scp Secure copy
script Make a typescript of a terminal session
sftp Secure file transfer program
sort Sort lines of text files
ssh RemoteHost Secure remote login
sudo cmd Execute command as Superuser (requires a password)
sync Flush filesystem buffers
tail file View last lines of a file
tracepath Trace path to a network host discovering MTU along this
path
traceroute RemoteHost Find the route to RemoteHost
uncompress expand compressed files
unzip List, test, and extract compressed files in a ZIP archive
uptime Tell how long a system has been running
vim Text editor (also vi)
vi Text editor (also vim)
wc Print the number of bytes, words, and lines in files
whatis Search the whatis database for complete words
whereis Locate the binary, source, and manual page files for a
command
which Show full path of shell commands
whoami Print effective userid
who Show who is logged on
whois Search the WHOIS database
w Show who is logged on and what they are doing
zip Package and compress files
/usr/sbin: Commonly used system commands

adduser
lpc
nfsstat
tcpdump
Heading
keyboard
definition key
key definition
Term . . . . . . . . . . . . . . . . . . . Glossary definition
command options description
<Super>: command # Superusr command
/home/joe user: command # User level command


Appendix C
Unix/Linux Commands by Category
The Filesystem
/ Root directory
. Current directory
.. Parent directory
Home directory
user Home directory
* Wild card: Match any sequence of characters
? Wild card: Match any single character
ls [D] List the contents of the working directory or directory D
ls -l List the contents of the working directory in long format
ln [opts] Existing New Create a link from New to Existing
cd Change working directory to home directory
cd .. Change working directory up one level
cd pathname Change working directory to pathname
pwd Path to the working directory
cp A B Create a copy of A named B
mv A B Move (rename) file A to B
mkdir A Make a directory named A
rm A Remove file A
rm -r * Remove (recursively) all files in working directory
rmdir D Remove directory D
du Report disk space used in the current directory and all
subdirectories
Compression and Archives:

compress file file.Z Compress a file
uncompress file.Z file Uncompress a file
gzip file file.gz Compress a file
gunzip file.gz file Compress a file
tar Tape archive untility
tar cf Archive.tar Dir Create an archive (Archive.tar) of the directory (Dir) and
its contents.
tar cfz Archive.tar Dir Create a compressed archive (Archive.tar) of the directory
(Dir) and its contents.
tar xf Archive.tar Dir Extract the tar file into the working directory.

APPENDIX C. UNIX/LINUX COMMANDS BY CATEGORY
tar xzf Archive.tgz Dir Extract the compressed tar file into the working directory.
tar tzf Archive.tgz Dir List the contents of the tar file.
tar xf Archive.tgz FileofInterest List the contents of the tar file.
Filesystem Management
du
df
dd
fdisk
mount
sync
umount
mkfs
mkswap
swapon

Printing
ers.
Processes and Multitasking

C-c Abort current task
C-z Suspend current task
bg Place suspened task in background
jobs List currentjobs processes
fg [JobId] Bring background task or suspeneded to foreground
task & Launch task into background
ps aux Report process status for all processes
kill -9 Pids Abort processes with Pids

top Display and update top CPU processes
nice Run a task with a modified scheduling priority
renice Alter priority of a running task
nohup cmd Run cmd as a background task immune to hangups.
crontab -e user Edit user’s cron file to schedule background tasks
Protection and Security

.nofinger File in users home directory created using the touch com-
mand limits finger to local machine
umask Set default file access permissions (permissions taken away
chmod [ugo][+|-][rwx] file Change file mode (permissions)
chmod octalmode Change file mode (permissions)
sftp RemoteHost Secure file transport, replacement for ftp
vim -x file Create encrypted file. You will be promped for an
encryption key
Networking Terms
TCP/IP Transmission Control Protocol / Internet Protocol
udp Best effort packet delivery protocol
tcp Reliable byte stream protocol
rpc Remote procedure call
ARP Address Resolution Protocol
route a network route to another machine
host Any machine on the network
client The host that wants data or a service
server The host that provides data or a service
Remote access and file transfers

telnet RemoteHost Remote login, often disabled for security reasons
rlogin RemoteHost Login to RemoteHost (insecure)
rsh RemoteHostcommand Execute command on remote host. Used in conjunc-
tion with .rhosts file. Often disabled for security rea-
sons.
ftp RemoteHost File Transfer Protocol transfers files between hosts
sftp RemoteHost Secure file transfer protocol
rcp fromfile tofile Remote copy (insecure)
scp file RemoteHost Secure copy file from here to RemoteHost
scp Host1:f1 Host2:f2 Secure copy f1 from Host1 to Host2
Email
.forward File in user’s home directory containing mail forward-
ing address
mail user[@RemoteHost] [¡ file ] Mail user agent
pine Mail user agent
elm Mail user agent
mutt Mail user agent
System Information:


hostid Report the numeric identifier of the local host
uname -a Report system information
uptime Report time since last reboot
du -h Estimate file space usage at and below the cwd
df -h Report file system disk space usage
ps aux Report process status
top Report top processes
Superuser/Root:
su [user] Switch to different user (user must be in /etc/passwd
file)
sudo command Execute command as root (User must be in
/etc/sudoers file)
chown owner : group file Change file/directory ownership
halt This and the next two commands reboot the system
(must be superuser)
reboot
shutdown
Ctrl-Alt-Delete Reboot the sytem
TCP/IP Diagnostics:
Some of these commands may require Superuser priviledges.
/etc/hosts Static table lookup for host names.
/etc/resolv.conf Static table with name of the DNS
/etc/services Internet network services list.
ifconfig Display or configure local TCP/IP network interface con-
figuration
ping RemoteHost Send a “ping” to a remote host to check connectivity
traceroute RemoteHost Print the route packets take to network host
ipmaddr
iptunnel
arp Display or modify IP-to-Physical address translation ta-
bles used by ARP
rarp
route Print out the local routing table (check /sbin/route)
dig RemoteHost Send domain name query to DNS
nslookup RemoteHost Send domain name query to DNS
tcpdump -i ethX Listen to network traffic on ethX
ethereal Capture and analyze network traffic
netstat Display protocol statistics
netwatch Monitor network connections
nmap Network exploration and security monitoring
whois Client for the whois directory service
Personal Information and Environment

.login File in home directory which Personalize the csh and tcsh environment. Read upon login.
.cshrc Files in home directory which personalize the csh, tcsh environment. Read upon login.
.profile File in home directory which personalizes the sh, bash environment
.forward File in user home directory containing email forwarding information

.signature Automantically added to your email
.plan File in user home directory containing public plans
.project File in user home directory
.nofinger If present denies existence of user outside of local host. Created by touch .nofinger
finger user[@RemoteHost] Lookup user information contained in password file and
the following files.
chfn Change use name and information in /etc/passwd file
source file Update environmental information from file (.cshrc).
System
su user Switch user (requires a password)
sudo cmd Execute command as Superuser (requires a password)
chown user:group file Change file ownership (must be Superuser)
df Report free space remaining in the file system
halt System shutdown
reboot System reboot
Ctl-Alt-Del System reboot
IO Redirection, Pipes and Filters

cmd < file Use file as input to cmd.
Example: mail user@RemoteHost< file
cmd > file Create file containing the output of cmd.
Example: cat file1 > file2 ; equivalent to cp file1 file2
cmd >> file Append output of cmd to file file
cmd1 | cmd2 Connect two processes with a pipe. The output of the
first is piped to the input of the second
Colaboration and Productivity Tools

cal [[month] year] Display calendar for month or year
date Display or set system date and time
w Display information on users currently logged on
who Display information on users currently logged on
whoami Display userid of current user
talk user@hostname Initiate a talk session with another user
write user@hostname Send a message to another user
msg y | n Enable, disable reception of talk and message from other
users
mail user@hostname ¡ file Send file as an email message to user@hostname
ispell file Look for misspellings in file and offer suggestions
aspell file Aspell (replacement for Ispell) looks for misspellings and
offers suggestions
look prefix Display words beginning with prefix
sort [rules] file Sort file according to rules
wc file List the number of characters, workds, and lines in file
diff file1 file2 List the differences between file1 and file2
cat list of files Concatentate files and sent to standard output
more file Display file one page at a time
less file Display file one page at a time and allow paging backwords
head file Display the first few lines of a file
tail file Display the last few lines of file


word)
ronment.

Bibliography
[1] Welsh, Dalheimer, and Kaufman, Running Linux, 3rd Edition. O’Reily & Associates, 1999.
[2] Ellen Siever, Linux in a Nutshell, 3rd Edition. O’Reilly & Associates, 2000.
[3] M. Sobell, A Practical Guild to Linux. Addison-Wesley, 1997.

[4] Oram, A and Talbott, S. Managing Projects with Make. O’Reilly & Associates, 1991.
[5] Vesperman, Jennifer, Essential CVS. O’Reilly & Associates, 2003.
[6] http://www.linuxdot.org/nlm/ Linux Newbie Guide.
[7] http://sunsite.auc.dk/linux-newbie/ Linux Newbie Administrator Guide.
[8] The AfterStep Homepage at http://www.afterstep.org

Study Notes GNU Linux Command Line Interface

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Study Notes GNU Linux Command Line Interface

Uploaded by

Copyright:

Available Formats

STUDY NOTES

GNU/Linux: the command line interface

DRAFT Version 0.9. Edited February 25, 2004

3 UNIX Command Line Basics 15

5 The Linux Environment 39

6 User Level Security 47

DRAFT COPY February 25, 2004 3

11 Networking and The Internet 75

12 Program Development in Unix/Linux 83

B Unix/Linux Commands by Location 95

C Unix/Linux Commands by Category 101

Linux is a computer operating system kernel originally developed by Linus Torvalds as a

1.1 Linux Features

• is a full-featured, 32-bit multi-user/multi-tasking OS.

• adheres to the common (POSIX) standards for UNIX .

DRAFT COPY February 25, 2004 5

1.2 Multi-User Operation

1.3 Why GNU/Linux?

Figure 1.1: Linux Uses

6 DRAFT COPY February 25, 2004

1.4 User Interface: Graphical User Interface vs Command Line

<Super>: indicates a command entered by the System Administrator or Root.

DRAFT COPY February 25, 2004 7

8 DRAFT COPY February 25, 2004

2.1 X Display Manager, Window Managers and Desktops

X-Window Managers for Linux

DRAFT COPY February 25, 2004 9

AfterStep NextStep feel and beyond

2.2 Starting and Customizing the X Environment

10 DRAFT COPY February 25, 2004

Figure 2.1: Gnome Interface

DRAFT COPY February 25, 2004 11

Figure 2.2: KDE Interface

12 DRAFT COPY February 25, 2004

# merge in Xdefaults and keymaps

# start some nice programs

DRAFT COPY February 25, 2004 13

14 DRAFT COPY February 25, 2004

UNIX Command Line Basics

3.1 Logging In To Your Account

DRAFT COPY February 25, 2004 15

You should see files like

drwxr-xr-x 13 joe user 1024 Nov 9 17:06 .dt

3.2 Command Structure

Special Keys Strokes:

16 DRAFT COPY February 25, 2004

Get Help and Information

Text Files and Printing

DRAFT COPY February 25, 2004 17

chown owner.group files Change file owner and/or group

3.3 The Linux Manuals and the man Utility

<bash>: man -k manual

18 DRAFT COPY February 25, 2004

LS(1) FSF LS(1)

Figure 3.1: Manual Page for ls

cd(n) Tcl Built-In Commands cd(n)

Figure 3.2: Manual Page for cd

DRAFT COPY February 25, 2004 19

3.4 File System: Create, List, Copy, and Move

3.4.1 Creating New Files

3.4.2 Creating New Directories

3.4.3 Listing Files and Directories

3.4.4 Copying Files and Directories