linux command with examples

December 11, 2012 by arunyadav4u 16 Comments

I am using Ubuntu for last 5 years and doing a lot of things from command line. Many of
the day to day command is now on my tip, but many of more I have used is out of my
mind, and I have to google it back, So I have decided to note it down here for my future
reference. I will keep adding to it whenever I get time or use new command.
Note :
Linux command can take a no of options. These options can be passed in both
shorthand form or expanded form. with shorthand form use single hyphen and with
expanded form use double hyphen -, say you can get help option to a command as -h
or -h
Some options do not need any argument while other need some argument to be passed
with them. you can pass all the options separately, but it is better to pass argument less
options grouped together and passing option with argument separately. Say, $ mkdir
-pv -m 777 myfolder is more concise then $ mkdir -p -v -m 777 myfolder. In the first
case we have grouped -p -v together as -pv. Since -m option need further argument
here we have passes 777, we have kept it separate from -pv
| , known as pipe is very important for filtering result when used with grep. | basically,
pass the result of first function as a block to the next function. Example, ls | grep
myfile will list only those files and folder which have myfile in there name. Here ls will
return list of all file and folders which will passed to grep, which will do the filtering.
1 -> history # will show you all the command you have typed in past
Syntax : history #it do not take any argument
Example:
$ history # will list all comand used in past
$ history | grep netbeans # since we have piped the result of history to grep it will
return only those command , which contain word history in them
2 -> man #show complete detail of a command
syntax : man command_name

It is basically short form of manual and will tell you detail about other commands you
pass to it as argument. It will explain you what that particular command do, how it work,
what options it will take and so on. Basically, if you remember name of any
command, you can get all detail of it with man command.
Example
$ man man # it will tell about how to use man itself
$ man sudo # will tell you all about sudo command
$ man dpkg # will tell you all about dpkg command
3 -> help #show tha available options for a command
Syntax: command_name -help or command_name -h
unlike man , it will just tell you about different options a command can take. Since, most
of the time we are most concerned about the available options a command can take
rather then the whole detail, It is used more often then the man command. Also, most of
the software , say netbeans comes with help option which will help you on how to
execute them. On other hand man will work only with ubuntu command.
NOTE : If you know name of a command, you can get all available options with -h or -help option. Remember that for some command -h not work but you have to pass -help. for example: rm -h will not work, you have to use rm -help to know about rm
options
Example:
$ cd /netbeans-6.9.1/bin #move to netbeans bin directory
$ ./netbeans -help # you want to start netbeans and want to know available options.
It has listed down all options for you
General options:
help
show this help
jdkhome <path>
path to Java(TM) 2 SDK, Standard Edition
-J<jvm_option>
pass <jvm_option> to JVM
cp:p <classpath>
cp:a <classpath>

prepend <classpath> to classpath

append <classpath> to classpath

4 -> ls # Listing files and folders

Syntax : ls [options] #options are optional, if you do not pass it will take the default, you
can see complete list of options

with ls -help
Examples:
$ ls #will list all folder and files but not hidden files or folder
$ ls -a # will list hidden files and folders also
$ ls -B #will not list the backup file ending with ~
$ ls -l # will do complete listing with all details like owner, date, permission etc. but will
not show hidden file
$ ls -l -a -B #will do all the above
5 -> mkdir # creating directory
Syntax: mkdir [options] directory_name #options are optional, if you do not pass it will
take the default, you can see list of options with mkdir -help
Example:
$ mkdir myfolder
$ mkdir -p myfolder1/myfolder2/myfolder3 # -p option tell to create nested directory
with parent child relation. Thus myfolder3 will be created in myfolder2 and so on
$ mkdir -p myfolder# the p option means parent , if the directory already exist do not
throw error but use the existing directory as parent
$ mkdir -pv -m 777 myfolder # it will not complain if directory exist, set the permission
to 777 and also print the task it performing on terminal
6 -> cd #changing directory
Syntax : cd directory_name # you can pass directory_name or .. or ~ to cd
Examples:
$ cd myfolder # take you inside myfolder
$ cd .. # take you out from current folder to the previous folder
$ cd ~ # take you out of every previous folder to the root of terminal

7-> touch # it will create a file if not present and if already present will modify its
modification or access time depending on argument you pass to it, but not overwrite its
content
Syntax : touch [options] file_name # touch help will show you all available option. If
you do not pass any option you can create multiple file with single touch command
Examples :
$ touch test1 #will create test1 file
$ touch test1 test2 test3 test4 # will create test2 test3 test4 as new file. Since test1 is
already there it will just modify its access time
$ touch -a test1 # will change access time of test1 file
$ touch -m test1 # will change modification time of file test1
$ touch -am test1 # will change both access as well as the modification time
$ touch -d '1 May 2005 10:22' test1 # with d option you can set any date you
want but maintain the formate of string, say '1 05 2005 10:22' otherwise

'2005 1 May 10:22' will give error. however, you can

ignore time if you want
$ touch -d '1 May 2005' test1 # it will set only the date, if you do not pass year then
it will take current year i,e '1 May' will also work
$ touch -r test2 test1 # -r is reference option i,e it will set datetime of file test2 to test1
file
8 -> rm # removing file or folder, by default it remove only files to remove a folder, you
need to pass additional options
Syntax: rm [options] file_or_directory_name #options are optional, if you do not pass it
will take the default, you can see list of options with rm -help
Example :
$ rm my_file # will remove my_file
$ rm -f my_file # will not throw error if my_file do not exist i,e it will ignore the removal
$ rm -fi my_file # Will ignore if file is not there and ask you to press Y to delete the file
i,e it will become interactive

$ rm -r my_folder # The -r option must be passed to delete a folder other while it will
not delete it
$ rm -rf my_folder # will delete the folder and will not complain if the folder do not exist
9 -> rmdir #it will list empty directory
Syntax : rmdir [options] directory_name # without argument it will delete all empty
directory name passed to it. It will throw error if directory is not empty. for that case use
rm as above
Examples:
$ rmdir dir1 dir2 dir3 # it will delete the empty directory dir1 dir2 dir3
$ rmdir -p dir1/dir2/dir3/dir4 # It will delete the parent also if deleting the child make it
empty, so first dir4 will be deleted then dir3 then dir2 and so on
10 -> mv # It will move file or directory to other location. if the location of source
destination is same the file or directory will be get renamed
Syntax: mv [options] source destination # If two files are provided as source and
destination source will be renamed as destination. If you pass more then two argument
and the last one is directory all the previous will be treated as source and will be moved
into the last directory, but if the last argument is not directory it will throw error.
Examples :
$ mv myfile.txt myfile.rb # it will rename myfile.txt to myfile.rb i,e the name is changed.
$ mv -bvi myfile.txt myfile.rb # here we are passing some option b will backup the file
before renaming, i will ask your yes no before renaming and v will print what going on
the console
$ mv test1 test_folder myfile.rb /home/arun/my_folder # it will move the file test ,
myfile.rb and the test_folder to /home/arun/my_folder
$ mv * /home/arun/my_folder # it will move everything in current folder to
/home/arun/my_folder
11-> grep # It is used to search a pattern specified by user in the given text or files or
block passed from first command to it through the pipe symbol |
Syntax: grep [options] pattern [files] # the complete list of options can be accessed with
grep -help command. If you want to match more than one word pass within

$ grep Linux is good file1 file2 file3 # It will list all lines of these three file containing
Linux is good in it in separate line, each line will be preceded with name of the file in
which they appear. If you do not want the file name you can use -h option as below
$ grep -h Linux is good file1 file2 file3 # It will not precede the matched line with
the file in which they are found
$ grep -i Linux is good file1 file2 file3 # it tell to ignore the case i,e linux is good
will also match
$ grep Linux is good * #will search the text Linux is good in all the text file in the
current directory
$ grep -r Linux is good * #-r option will make it to search the text Linux is good in
all the text file recursively in the current directory as well as all it child directory
$ grep -c Linux is good file1 file2 # It will not print each line but just return the count
of match in each file
$ grep -n Linux is good file1 file2 # It will also print the line number of the line
containing the pattern in each file
$ grep -rni Linux is good * # will recursively search each directory, print the line no
and will ignore the case
$ ls -l | grep foo # it will list only those file and directory which containn the word foo
$ history | grep rails # will list history of only those command which contain the word
rails
$ history | grep -in rails #will ignore case, and print line no.
12 -> less # can be used to see content of a file .It allow you to display a large data in
block and let you to scroll down, basically, it keep the screen at first line and let you
scroll down. If you do not use this, screen will set to the last line and you have to scroll
up to see the first line. you can exit the content by pressing q .
Syntax : less [options] file_name # you can see complete list of option with less -help.
Though, generally, I used it to see the content of file or a command from top and have
not tried any options .
Examples:
$ less myfile.txt # It will show content of file from the top, need to scroll down to see
more content. press q to exit the content

$ ps -ef | less #will show the result of ps -ef command from the top
13 -> more # It can be also used to display content of a file on terminal.
syntax: more [options] file_name # By default It do not provide scroll down facility like
less to see the whole content but show only that much content which fit into your
screen. however, using different options you can set the no of line you want to see,
which will activate the scrolling if the set line not come on the screen
Examples:
$ more myfile # will show that much content of file which fit in screen
$ more -100 + 50 myfile # will show 100 lines of file starting from line 50
$ more +/Iam arun -100 # will show 100 line of the file from the line which contain
Iam arun
$ more -s -u -100 +50 myfile # s will cause to remove blank line from display and u
will remove underlines
$ ps -ef | more -10 + 20 # will display 10 lines of the result of command ps -ef starting
from the 20 line
14 -> cat # will be used to create a file, read a file, concenate content of many files
syntax : cat [options] [input_filename] [> or >>] [output_filename] # all the things in [] is
optional i,e cat can work without them also
NOTE :
1 -> If you do not provide output_filename the output will be printed on terminal itself i,e
terminal is the standard output for cat command
2 -> > and >> both are called append operator i,e write content to a file but > will erase
the file of content if exist and write to it but >> will add the new content to the existing
content , so it is safer to use >> instead of>
Example:
$ cat >> test_file1 # it will create test_file1 and when you press enter it will take you to
the next line where you write to it, to exist press enter then press ctrl and d
I have pressed enter on last line to reach here
again pressed eneter to come to next line and keep writing
All content Iam writing will get written to test_files1
to exist writing we will press Enter followed by Ctrl D

$ cat test_file1 # it will show you content of file test_file1, so you can see the content
you have written on the terminal
I have pressed enter on last line to reach here
again pressed eneter to come to next line and keep writing
All content Iam writing will get written to test_files1
to exist writing we will press Enter followed by Ctrl D
arun@arun-yadav:~$ cat -n test_file1 #the -n option will print the line no also
1 I have pressed enter on last line to reach here
2 again pressed eneter to come to next line and keep writing
3 All content Iam writing will get written to test_files1
4 to exist writing we will press Enter followed by Ctrl D
$ cat >> test_file2 # will create a new file
This is the second file # you added this line to test_file2
$ cat test_file1 test_file2 >> test_file3 # will create file test_file3 and append content
of test_file1 test_file2 to it
$ cat test_file3 # it show the content of test_file3, you can see that this file contain
content of both the files
I have pressed enter on last line to reach here
again pressed eneter to come to next line and keep writing
All content Iam writing will get written to test_files1
to exist writing we will press Enter followed by Ctrl D
This is the second file
15 -> echo # you can use it to write a string to file or terminal itself
syntax: echo string [filename] # file name is optional if you do not pass it the output will
be printed to the terminal
unlike cat, it can use to write single string to a file or terminal i,e you can write multiple
lines as with cat command
Example:
$ echo Iam Arun # it will print the output to terminal
Iam Arun
$ echo Iam Arun to test_file >> test_file # it will append the string to the test_file, if
test_file is not present >> operator will create it and the append to it
$ cat test_file # we will see the content of test_file
Iam Arun to test_file
$ echo Iam adding one more line >> test_file # one more line is appended
$ cat test_file # you can see both the line
Iam Arun to test_file

Iam adding one more line

$ echo this > operator is dangerous as it will erase and write > test_file # see the
danger of using >, if the file exist it will erase its content before appending
arun@arun-yadav:~$ cat test_file
this > operator is dangerous as it will erase and write # see earlier line is erased and
only single line present
16 -> pstree # will return all the running process in tree structure with there name and
process id
Syntax : pstree [options] [pid or username] # you can see the complete list of option
with pstree -help. the argument is user or process_id whose tree you want to see
Example:
$ pstree #will list all the running process with all the user
$ pstree root # will show all the process running with root user
$ pstree arun # will list all the process running with the user arun
$ pstree -p arun # will print the PID i,e process id also with the name
$ pstree -pn # will print the PID and also in the sorted order
$ pstree -pnh # the h option will highlight the current running process
$ pstree -pnhu # the u option will also print the user owning that process
$ pstree -pnhul # l option will not allow to truncate the process detail if it is too long
17 -> pgrep # It will return PID of the process
Syntax : pgrep process_name # it will return the PID of process_name. It will be helpful
if you know the process name otherwise use pstree or ps to find PID
Examples:
$ pgrep firefox # will return PID of firefox
$ pgrep ruby # will return PID of ruby
$ pgrep skype # will return PID of skype
18 -> ps # it means process status and return all the running process with name and
PID

Syntax : ps [options] # you can see all the available options with ps -help
Examples :
$ ps # will return currently active process
$ ps -e # will return all the running process
$ ps -ef # will return all running process with full detail
$ ps -ef | less # as process list is very long, we have piped the output of ps -ef to less
command. It help you to display result in small blocks and let you to scroll down to see
more.
NOTE : IF you want to kill some process through script, you will jut need the PID to pass
it to kill command. say to kill apache 2 from script you can use below:

kill -9 $(ps -e | grep apache2 | awk '{print $1}')

19 -> pkill # will kill the process passed to it
Syntax: pkill process_name # It would be helpful only if you know the name of process,
you want to kill
Examples:
$ pkill ruby
$ pkill firefox
20 -> kill # will kill the process whose PID is passed to it
syntax : kill [signals] PID # there is 60 type of signal which can be passed to kill, but
you have to just remember 9 which is called SIGKILL, the default signal passed is 15
which is called SIGTERM. Actually, kill do not kill the process itself instead, it just pass
proper signal to a process which cause it to terminate i,e logic of termination is in every
process itself. Say, for example when you close firefox browser, signal 9 get passed to
the firefox process by our OS and it get terminated. The kill command just give you a
way to pass signal to a process from the terminal.Thus by passing proper signal with kill
you can control any state of a process. But mostly that is not needed. Mostly it is use to
close any hanged process. If any process get hanged, you get its PID from the above
pstree or pgrep or ps command and the pass it to kill to terminate that hanged process.
NOTE: the complete list of signal available for kill can be listed with kill -l
Examples:

$ kill 485 #will terminate the process with PID 485. here the default signa 15 get
passed, there may be chance that it fail to kill the process, say if it is making some
system call
$ kill -9 485 # it will be guaranteed that the process will be called. If the process is
making any system call, it will wait and kill it when it return from the call
21 -> killall #will kill the process along with its child whose name passed to it. helpful if
you know the process name
syntax: killall [options] process_name #will kill the process and all its child
NOTE: Killing by file name only works for executables (i.e., runnable programs)
that are kept open during execution
Examples:
$ killall nautilus #will kill the nautilus process
$ killall -e nautilus # it will tell to kill whose name exactly match with nautilus. It will be
helpful if your process name is more that 15 character, since killall by default match only
upto that, so will delete all other process also whose first 15 character matches. the -e
option will prevent this.
$ killall -ew nautilus # the w option will tell to wait till all the process are killed,
otherwise killall check it every second by default and resend the signal if found some is
still running, but since some has already get killed, some signal become unresponsive.
Also killall can kill all the process, including other killall process, except itself
22 -> pwd # will print the current working directory
23 -> clear # It will remove all previous command and output from the console
24->SCP # You can take it as server copy. scp command is used to copy files from one
server to other.It can be from your local machine to some other machine or vice versa.
You can also use it to copy file between two different machine from your system.
Below will be the syntex :
CASE 1: Copy file from from your system to some other system:
scp /path/to/local/file username@hostname:/path/to/remote/file
CASE 2: Copy file from another system to your system:
scp username@hostname:/path/to/remote/file /path/to/local/file

CASE 3: Copy file from some system to some other system from your machine:
scp username1@hostname1:/path/to/file
username2@hostname2:/path/to/other/file
NOTE : to copy a directory instead of file in any of the above case use scp -r instead
of just scp
NOTE : username@hostname is basically the url to which you ssh
24-> lscpu
This will tell you information about your CPU. Like you may want to know, you are
having 32 bit or 64 bit processor (remember many a time at the time of download you
do not know you should download 32 or 64 bit version).
Example :
$ lscpu
Architecture:
CPU op-mode(s):
Byte Order:
CPU(s):
On-line CPU(s) list:
Thread(s) per core:
Core(s) per socket:
Socket(s):
Vendor ID:
CPU family:
Model:
Stepping:
CPU MHz:
BogoMIPS:
Virtualization:
L1d cache:
L1i cache:
L2 cache:
L3 cache:

i686
32-bit, 64-bit
Little Endian
4
0-3
2
2
1
GenuineIntel
6
42
7
3092.905
6185.90
VT-x
32K
32K
256K
3072K

Architecture showing to be i686(X86, i686, or i386 is same), which means

your OS is of 32 bit. But the CPU op-mode showing that it can work in both 32 and 62
bit mod i,e your CPU is of 64 bit. So if you want you can install 64 bit Ubuntu OS on it,
but if the mode is only 32 bit, you cant
NOTE :

X86, i686, or i386

means you are running a 32 bit kernel i,e OS.
X86_64 , amd64 , or X64 means you are running a 64 bit kernel i,e OS.
24 -> netstat
It display information about all the running internet connection be default. if you pass
-a option it will display inactive connection also. It take a number of options. You can
find all the available option with below command:
# netstat -h
-r, route
display routing table
-i, interfaces
display interface table
-g, groups
display multicast group memberships
-s, statistics
display networking statistics (like SNMP)
-M, masquerade
display masqueraded connections
-v, verbose
-W, wide
-n, numeric
-N, symbolic
-e, extend
-p, programs
-c, continuous

be verbose
dont truncate IP addresses
dont resolve names
resolve hardware names
display other/more information
display PID/Program name for sockets
continuous listing

-l, listening
display listening server sockets
-a, all, listening display all sockets (default: connected)
-o, timers
display timers
-F, fib
display Forwarding Information Base (default)
-C, cache
display routing cache instead of FIB
You can use any of the above options
Example:
root@arun-desktop:~# netstat -ltnp
Active Internet connections (only servers)
Proto Recv-Q Send-Q
Local Address
PID/Program name
tcp
0
0
127.0.0.1:631
2322/cupsd
tcp
0
0
0.0.0.0:64283
3176/skype
tcp
0
0
0.0.0.0:902
1135/vmware-authdla
tcp
0
0
127.0.1.1:53

Foreign Address

State

0.0.0.0:*

LISTEN

0.0.0.0:*

LISTEN

0.0.0.0:*
0.0.0.0:*

LISTEN
LISTEN

1460/dnsmasq
tcp6
0
2322/cupsd
tcp6
0
10916/apache2
tcp6
0
10916/apache2
tcp6
0
10916/apache2

::1:631

:::*

LISTEN

:::443

:::*

LISTEN

:::8140

:::*

LISTEN

:::80

:::*

LISTEN

25-> browsing folder as root user.

sometime you want to open folder as root user, so that you can directly browse and edit
a file. For example, say you want to edit standalone.xml file of jboss . If you go to that
file through the UI, open it and edit it, you cant as you have browsed it as normal user
so dont have permission to edit it. So in such case you can browse the file as root user
with below command.
root@arun-desktop:~# nautilus
arun@arun-desktop:~$ sudo nautilus # for non root user