All important Linux commands

Like most modern operating systems, Linux has two available interfaces for user input. All of the settings that you set via the graphical user interface (GUI) can also be made in the form of command line directives via the so-called shell.

Shell is a program that functions as an interface between system and user. It includes a command line interpreter that accepts user input via the keyboard, evaluates them, starts programs (if necessary), and returns the output in the form of a text entry to the user. In addition, each shell has its own programming language which makes it possible to write shell scripts – for example, to link program calls and facilitate administrative tasks.

Each shell runs in a terminal. At the beginning of the computer age, independent devices, or so-called hardcopy terminals (printer or screen plus keyboard), were used. These were replaced on modern computers by terminal emulators – programs that provide users with a graphical window for interacting with the shell.

As soon as you access the terminal of your operating system, is starts the standard shell (i.e. the Bourne again shell, Bash) specified in the settings, and accepts input at the prompt.


Over time, various shells for unix-like operating systems have been developed which differ in terms of functionality and user-friendliness. As a Linux user, you have the choice of which command line interpreter to use. On most operating systems, multiple shells are already installed. The switch from one shell to another can be comfortably carried out from the terminal (see chsh in the user account management chapter). Among the most popular shells, beside the standard programs Bash and Dash, are Fish, Z-shell, Korn-shell, (t)csh, and Mksh.

$1 Domain Names

Register great TLDs for less than $1 for the first year.

Why wait? Grab your favorite domain name today!

Matching email
SSL certificate
24/7/365 support

Interaction with the shell usually happens via commands, which can be used to call command line programs of the same name. For every action that you want to carry out via the terminal, use a program call following this basic scheme:


A program call via the terminal uses the name of the program. Most programs offer the possibility to address certain program functions via options. If a program expects arguments – i.e. in the form of files or index paths – these are usually specified according to the selected options.

In the following sections, we offer an overview of the most common Linux commands and their associated command line programs.

Basic commands

In the basic commands category, you’ll find the Linux basic commands that are used to control the terminal. Learn how to clear the terminal’s visibility, retrieve previous terminal entries from the history, or exit the terminal session.




Clear terminal

Use the command line directive clear to clear the screen content.


You’ll receive a blank terminal with a prompt. Older entries remain in the scrollback buffer.

Instead of using this command, you can also clear the terminal with the key combination [Ctrl]+[L].


End session

The command line directive exit end the current session and closes the terminal.


Instead of this, you can use the key combination [Ctrl]+[D].


View list of all shell commands

Use the help command to see a list of all integrated shell commands (built-in commands).


Call help in combination with a shell command to retrieve a short description of the demand in question.



Release history file entries

In Bash, the last 500 commands entered in the command line are saved in the history.

This function serves as entry assistance, and allows you to look through the list of previous commands using the arrow keys and press ENTER again to confirm.

The history can be searched using keywords with the key combination [Ctrl]+[R]. You also have the option to view the complete list, numbered in the terminal.

Use the command history without options and arguments.


If you want to filter the results, combine history via Pipe with the command line programme grep (see search options) and a keyword.

history | grep KEYWORD

Help pages

You don’t know what to do? No worries. Under Linux, there are various help and documentation pages available directly via the terminal, such as the Unix man-pages and GNU info pages. These contain a detailed description of all command line programmes, system calls, configuration files, file formats, and core functions. With whatis and apropos, you can find command line programmes in the help pages category, which allow you to search the manual pages of your operating system for keywords.




Search the manual

Use apropos to search the page titles and descriptions of your operating system’s manual by keywords. The command line programme gives you all matches, including a brief description in the terminal.

Refer to the following scheme:


The command supports different options. Use the –e (--exact) option to limit the search to exact matches, or use wildcards (-w ‘*KEYWORD’) and regular expressions (-r).


Call the GNU info page

Via the info command, you can retrieve the GNU info pages for a specific topic. In most cases, these pages correspond to the manual pages that can be accessed via man, but as opposed to these, they have links that make the navigators in the manual simpler to read.

Use the following syntax to call a GNU info page:


A call without an option or topic leads you to the main menu of the GNU info page.


Call the manual

The man command opens the manual pages (man-pages) of your Linux distribution directly in the terminal.

Use the following scheme to call the manual pages:


The Linux man-pages are divided into 10 topic areas:

(1)    User commands

(2)    System calls

(3)    Functions of the programming language C

(4)    File formats

(5)    Configuration files

(6)    Games

(7)    Miscellaneous

(8)    System administration commands

(9)    Core functions

(10)New commands

For example, if you want to open the manual page for a specific Linux command, use it in combination with the name of the command.

man clear

You can also narrow down the search by specifying the topic area number:

man 1 clear

In both cases, the manual page opens to the command line directive clear. Use the [Q] key to close it and return to the prompt in the terminal.

The apropos command provides a way to search the Linux manual pages by keyword.


Call Lynx-style info pages

With pinfo, you have a variant of the command line programme info, which is based on the command line browser Lynx and issues information pages with highlighted links.

Use pinfo the same way as the info command:



Search the manual pages by keyword

The command line programme whatis serves as a keyword search in the manual pages. Call this programme with a popular keyword to search your operating system’s manual for exact matches. If there is a match, whatis gives a brief description in the terminal.


whatis (-w ‘*KEYWORD’) also supports placeholders and regular expressions (-r).

Directory operations

You’ll use Linux commands for directory operations to create, delete, and manage directories on your system through the terminal, as well as navigate the directory tree. The most important command line directives in this category are cd, ls, mkdir, and rmdir.




Navigation in the directory tree

The command line directive cd stands for change directory, and is used for navigation in the directory tree.

The syntax of the command reads:


If no target directory is given, then cd automatically switches to the user’s home directory.

If cd is used with a minus symbol after it (-), it changes back to the previous directory.


Execute programme in a new root directory

The chroot command (short for change root) is used to execute a command in a different root directory. For example, chroot is used to isolate critical programmes from the rest of the file system. In this case, it’s referred to as chroot jail.

Calling the programme requires root privileges, and is based on the following formula:



List directory content

The command line directive ls stands for list and is used to display the content of a directory (the names of all files and folders found in the given directory).

The syntax of the command reads:


If ls is used without a directory entry, then the command lists the content of the current directory.

With the help of additional options, you can define which information is shown and how it’s displayed.


Create directory

The command line directive mkdir stands for make directory, and allows Linux users to make new directories.

Use the following syntax to create a new directory in the current directory:


You can also create multiple directories at the same time by placing the names separated by a space:


If a directory is supposed to be created in a particular target directory, then specify the absolute or relative path to the directory.

mkdir /home/user/Desktop/test

mkdir ../Desktop/test

With both examples, the test directory will be created in the desktop directory.


Create directory hierarchy

With mkdirhier you can create entire directory hierarchies with a single command line directive:

mkdirhier [OPTION] /home/user/directory1/directory2/directory3

If directory1 and directory2 already exist, mkdirhier only creates directory3. Otherwise, all three directories are created.


Output directory name

Use pwd (short for print working directory) to output the name of the current working directory.

The syntax of the command reads:



Delete directory

If you want to delete a particular directory, use the command line directive rmdir (remove directory) according to the following syntax:


You can only delete empty directories with rmdir. To delete a directory along with all of its contained files and subfolders, use the command rm (remove) with the option –r.

Warning: rmdir doesn’t require a confirmation for deletion. Selected directories are irreversibly deleted.


List directories in the tree:

While ls only lists the content of a directory, the command line directive tree can be used to output the entire directory hierarchy recursively as a tree structure.

The command uses the following syntax:


File operations

The Linux commands in this chart allow you to carry out various file operations from the terminal. Use the Linux basic commands like cp, mv, and rm to copy, move, rename, or delate files on your system.




Output file name

A file path is passed to the command line directive basename, which simply returns the file name without a default path.

The syntax of the command reads:

basename [OPTIONS] path/to/files [SUFFIX]

For example, if you enter $ basename/home/user/photo.jpg in the terminal, you’ll receive the following output:


The additional input of the suffix removes this from the output as well.

Input: $ basename/home/user/photo.jpg .jpg

Output: photo

The command can be expanded to multiple files using options.


Combine file content

The command line programme cat (short for concatenate) was developed as a tool for the combination of file content and can be used as a pager for the display of file content in the terminal.

Use cat with the following syntax in the terminal to read a file and output it to stdout (the standard output):


Multiple files can be separated by spaces:


Linking file content is done with the use of redirection operators (>, <, and |). For example, use the operator “bigger than” (>) to combine the content of two files into a third.

cat file_1.txt file_2.txt > file_3.txt


Align files at byte level

cmp is part of the diff package and is used to compare file contents. As opposed to diff, the alignment is done at the byte level and so is particularly suitable for binary files.

Use cmp according to the following syntax:


cmp finds differences, and then the command line programme outputs the byte and line number of the first deviation in the terminal.


Compare sorted files by the line

Use the command line programme   comm to compare sorted files (i.e. via sort) line by line.

The programme call is based on the following syntax:


If come is called without options, the programme generates an output with three columns: The first column contains all lines that only appear in FILE1, the second column contains all lines only in FILE2, and the third column contain all lines that appear in both files.

The programme supports three options:

-1: suppress unique lines from FILE1

-2: suppress unique lines from FILE2

-3: suppress all lines contained in both files


Copy files or directories

The command line directive cp (copy) is used to copy files and directories. The basic syntax of the command reads:


The SOURCE is the element that is intended to be copied. Either a file or a directory is then defined as the TARGET of the copying process. If you define an existing file as the target file, its contents are overwritten with the source file. You also have the option to create a new file with whatever name you choose as the target file.

If you want to copy multiple files, then the target has to be a directory. The same goes for copying a directory.

Copy a source file to a target file in a current directory:


Example: cp file.txt file_copy.txt

Copy a source file from the current directory to a target directory:


Example: cp file.txt home/user/documents/2017

Copy multiple source files to a target directory:


Example: cp file.txt file.odt home/user/documents/2017

Copy a source directory from the current directory to a target directory:


Example: cp directory1 home/user/documents/2017

If a directory is meant to be copied along with all its contents, then all subdirectories need to be included in the copy process using OPTION –R.


Extract file content

The cut command allows you to extract the contents of a file from the text line of a file (i.e. log or CSV files).

The syntax of the command reads:


The exact position of an extracted section is defined via the options –b (byte position), -c (character position), -d (delimiter), and –f (field).


Compare files or directories

The command line programme diff serves to compare two files. You can also use diff to determine if two directories contain the same files.

Call the programme to the terminal using the following syntax:



Output file path

dirname is the counterpart to basename. The command line directive allows you to extract the path portion from a file path and output it in the terminal without file names.

The syntax of the command reads:

dirname [OPTIONS]

For example, enter $ dirname/home/user/photo.jpg into the terminal to get the following output:



Output file type

With the command line directive file you can output information about the file type of a file.

The call is based on the following syntax:



Create a shortcut to a file or directory

The command line programme ln (short for link) generates a shortcut to a file or a directory. This creates another directory entry for this file, which allows you to access the respective file via another file path.

The call for the command line programme has to always contain at least the path to the source file.

ln [OPTIONS] path/to/sourcefile

In this case, a shortcut will be created in the current work directory under the same name.

You can also enter a target path and then name the shortcut however you want:

ln [OPTIONS] path/to/sourcefile path/to/shortcut

By default, ln creates so-called hardlinks. These aren’t suitable for creating shortcuts to directories. Hardlinks also can’t be used beyond partition boundaries. So the command is often used with OPTION –s (--symbolic), with which symbolic links can also be created beyond file system boundaries. Symbolic links also point to and depend on a “real” file path.


Output open files in the terminal

lsof stands for list open files, a tool that gives you information about open files in the terminal, sorted by PID (process ID).

Call the programme to the terminal using the following syntax:

lsof [OPTIONS]

Since unix-like systems such as Linux generally follow the policy that “Everything is a file,” the list outputted by the lsof command is accordingly long. As a rule, the options are used to limit this output.


Calculate checksums

The command line directive md5sum helps you calculate and check MD5 checksums for files.


Move file or directory

The command line programme mv (move) copies a file or directory and deletes the original element. If it’s used within the same directory, then mv can be used to rename files.

The programme   call is based on the following syntax:


Application examples:

Move a file to another directory:


Example: mv file1.txt home/user/documents/2017

Move multiple source files to a target directory:


Example: mv file1.txt file2.txt home/user/documents/2017

Move a subdirectory from the current directory to a target directory:


Example: mv directory1 home/user/documents/2017

Rename a file in the current directory


Example: mv file1.txt file2.txt

Rename a subdirectory in the current directory:


Example: mv directory1 directory2


Combine file contents by column

Similar to cat, the command line programme paste also enables the output of file contents to the standard output. But while cat merely combines content, paste joins column by column.

The basic syntax of the command reads:


In standard mode, the listed files are merged so that all rows with the same row number are transferred to the same line in the output. Every line in the output contains content from all input files.

You can customise which separator is used by paste with the option -d. Tabs are used as the default separator.

A second mode can be activated using the -s option (serial). With this, all lines of the first input file are transferred to the first line of the output. The data for all other input files follows in separate output lines, so each line of the output contains the contents of only one input file.


Rename files

The command line programme rename enables the renaming of files and folders with the help of Perl-compatible regular expressions (regex). As opposed to mv, the rename function is suitable for file operations where the names of several files are supposed to be either partially or completely adapted.

Use rename according to the following syntax:


The syntax of regular expressions reads:


The following example renames all .html file endings to .xhtml.

rename 's/\.html$/.xhtml/' *.html


Delete file or directory

The command line programme rm (remove) permanently deletes files or entire directories.

The programme call is based on the following syntax:




If a directory is to be deleted along with all its subdirectories, then use rm plus the option -R (--recursive).


Multiple files or directories are separated by spaces.



“Shred” files

shred is a command line programme that enables safe deletion of files. Chosen elements are overwritten in the course of the deletion process and so can’t be restored by forensic means.

The overall syntax of the command reads:


Use shred with the following options to permanently delete a single file:

shred -fuz FILE

The -f option forces the delete process, -z overwrites the file contents with zeros (default is random data), and then finally -u removes the shredded file from the file system, similar to the rm command.


Sort file lists and programme output

Use the command line directive sort to sort file lists and programme output numerically, alphabetically, and by row.

The overall syntax of the command reads:


The sorting method can be customised using options: For example, numerical (-n), random (-R), or in reverse order (-r).


Split files

The command line directive split is used to divide files.

The underlying syntax reads:


The placeholder INPUT corresponds to the file that is to be split. The PREFIX acts for the names of the participating files. Their name is based on the following pattern:



Output time stamp

The command line directive stat (status) outputs access and alteration time stamps for selected files and directories.

The general syntax of the command reads:


The output format can be customised with the use of options.


Change time stamp

The command line directive touch can be used to modify access and alteration time stamps for files. If touch is applied to a file that doesn’t exist yet, then it is automatically created. So the command is also good for creating empty files.

Use touch according to the following pattern:


To set the time stamp for a file to the desired date, use the OPTION -t along with the time information in the forms [YY]MMDDhhmm[.ss].


touch -t 1703231037 file.txt

Access and alteration time stamps are now set to March 23, 2017, 10:37. The modification can be restricted to access or time stamps with the options -a and -m.

If the touch command is used without option -t, then it uses the current time stamp.


Delete duplicates in file lists and programme outputs

The command line directive uniq is usually used in combination with sort to clean sorted files from duplicate lines.

In the following example, the sort command is linked by a pipe (|) to the uniq command to first sort a file and then output it without duplicate lines.

sort file.txt | uniq

Rights management

It’s quite easy to customize access and ownership rights for files and directories in the terminal on Linux. The most important command line directives for rights management are chown and chmod. Group affiliations are managed by the chgrp command.




Manage file attributes

The command line program chattr (short for change attribute) allows you to view files or directories with attributes. An adjustment of file attributes is supported by various file systems (i.e. ext2, ext3, ext4, XFS, ReiserFS, JFS, and OCFS2).

Use chattr according to the following syntax to set an attribute:


Set attributes can be removed again using this pattern:


For example, set the attribute -i to prevent changes (deletions or modifications) to a file or a directory:

chattr +i file.txt

For other attributes and possible options, refer to the chattr program manual.


Manage group affiliations for files and directories

The command chgrp stands for change group and is used for the management of group affiliations for files and directories. To be able to use chgrp on a chosen file or directory, you have to have owner or root permissions. These are the only groups to which you can belong.

chgrp is used according to the following syntax:




The option -R refers to subfolders and files contained in a directory.


Manage access rights

The command line program chmod (short for change mode) is used to assign rights in unix-like file systems (i.e. ext2, ext3, ext4, reiser, xfs).

The general syntax of the command reads:




The MODE placeholder stands for the applicable rights mask. You can find out more about how to create such a system and what to pay attention to in our guide on access rights with chmod.

With the help of the -R option, rights can be assigned recursively to subfolders and files contained in a directory.


Manage ownership rights

Usually, the creator of a file or directory is automatically its owner. The command chown stands for change owner and allows you to modify the owner permissions.

The command is used according to the following syntax:




To set owner rights for a user or group, there are four possible combinations available.

Owner and group are reset according to the input:

chown [OPTIONS] owner_name:group_name file.txt

The group is reset according to the input, the user remains unchanged:

chown [OPTIONS] :group_name file.txt

The owner is reset according to the input, the group remains unchanged:

chown [OPTIONS] owner_name file.txt

The user is reset according to the input. The group is set to the default group for the logged-in user:

chown [OPTIONS] owner_name: file.txt

The changes are recursively extended to subdirectories with the help of OPTIONS -R.


Display file attributes

If you would like to display which attributes are set for a file or directory, use the command line directive lsattr (short for list attributes) according to the following syntax:


Search options

Linux offers various command line directives for searching through the system directly from the terminal.




Search file system

Using find opens up a command line program that serves to search through files.

The program call is based on the following syntax:


The specified directory is the starting directory of the search. The command then searches the starting directory and its subdirectories. If no directory is entered, then find starts the search from the current working directory.

Options allow you to define search criteria and actions. The default action is preset at -print: The output of the complete file names of all search results to the standard output (usually the terminal).

Common search criteria include the file name (-name FILENAME[SUFFIX]), a username (-user USERNAME), the file size (-size n[cwbkMG]), the access time in days (-atime [+-]n), or the modification time in days (-mtime [+-]n).

Search by file name uses meta-characters and placeholders. Put these in quotation marks to prevent interpretation by the shell.


find /tmp -name “*.odt” -mtime -3 -size +20k

The /tmp directory is defined as the starting directory. The command line program find outputs all files to the standard output that contain the .odt file ending, are larger than 20k, and were changed for the last time less than three days ago.

For other options for the find search command, refer to the program’s manual.


Search text files

With the grep command (short for global regular expression print), you can search through text files (i.e. log files). Any character strings or regular expressions can be used as a search pattern.

Use grep according to the following syntax:


If grep encounters a string that corresponds to the search pattern, then the line number along with the file name is output to the terminal.

In general, grep is used on all files in the current directory. The option -r enables a recursive search into the subdirectories.


Search file index

The command line program locate also allows you to search for files through the terminal. But as opposed to find, instead of searching through the file directory, it searches a specially created and regularly updated database. As a result, locate provides results must quicker than find.

To search the database for a particular file, locate is used according to the following syntax:


The search pattern can contain meta-characters as placeholders (*). Put these in quotation marks to prevent interpretation by the shell.

In the following example, locate outputs all files with the .png ending.

locate “*.png”

The locate command is case-sensitive. To gloss over the difference between upper-case and lower-case letters in the search, use the option -i.

The file /var/lib/locatedb functions as a database for the indexing of files. This contains a list of all the files of the file system at a particular time, and so must be regularly updated. To do this, use the updatedb command.


Vague search in text files

tre-agrep also is used to search for strings in text files based on search patterns. But unlike grep, it’s not only exact matches that are output, but vague results are also allowed, such as those with transposed letters or missing characters. The program is based on the TRE library and makes it available in the command line.

The syntax of tre-agrep matches that of the grep command:


Using options, you can define a maximum error allowance. In the following example, a maximum of one deviation is tolerated.

tre-agrep -1 'Linux' test .txt

tre-agrep outputs all lines of the text file test.txt that contain the words defined by the search pattern or which deviate from it by one letter: i.e. Linus.


Update file index

A locate search only functions properly if the /var/lib/locatedb file is continuously kept up to date. The updatedb command allows you to manually update the database. Note that you need root permissions to do this:



Search the binary code, source code, or manual of a program

With the whereis command, you can locate the binary code, source code, or manual files of the selected program.

The general syntax of the command reads:



whereis firefox

For the output, whereis writes the path to the found files separated by spaces in the terminal:

firefox: /usr/bin/firefox usr/lib/firefox usr/share/man/man1/firefox.1.gz

Options can be used to limit the search to specific file types or directories.


Identify program binary files

If you would like to identify the binary files of a program, use the command which with the following syntax to output the path in the terminal.



which firefox



In the default mode, which outputs the first file it finds. Use the option -a to show all files that fulfill the search criteria.

User information

Use the command line programs for the following categories to access detailed information on the registered users in the system as well as their groups and processes.




Obtain user information

The command line program finger serves to access user information. Use the command in combination with the desired username:

finger [options] [USERNAME]

The output contains the following information for the given user account: login name, real name, login time, time since last activity (idle time), home directory of the user, login shell, location (office number), mail and telephone number (if given), and the contents of the files .plan, .project, .pgpkey, and .forward in the home directory of the user (if given).

Use finger without a username to obtain information about your own account.


Obtain group affiliations

The command groups lists the group affiliations of a selected user account.

Use the command line directive according to the following pattern:


Use groups without a username to list all groups to which your user account belongs.


Obtain user and group IDs

The command line directive id outputs user and group identifiers of the selected user accounts.


If you want to identify your own IDs, use the command without a username.

The range of the output can be limited using options.


Obtain information about recently logged in users

Use the command last according to the following pattern to view a list of recently logged-in users, including login and logout times.

last [OPTIONS]

The corresponding information is obtained from the wtmp file under /var/log/wtmp. If you only want to request information about a particular account, then enter the command line directive with the desired username.



Obtain current user and their processes

The command w outputs a list of all registered users including all processes that they’ve executed.

Use w in combination with a username to limit the command to just this user:


Range and format of the output can be customized using options.


Obtain detailed information on registered users

The command who outputs detailed information about users registered on the system.

The general syntax of the command reads:


who supports various options with which you can customize the range of the outputted information.

By default, who refers to data about currently registers users from the /var/run/utmp file.

You have the option to specify one of the following files as the source of the information.

If you would like to obtain information about earlier registrations, use who in combination with the source file /var/log/wtmp:

who [OPTION] /var/log/wtmp

For information about unsuccessful registrations, enter the command along with the /var/log/btmp file:

who [OPTION] /var/log/btmp


Obtain your own username

Use the command whoami to obtain your own username.

whoami [OPTIONS]

User account management

Linux provides you with a series of programs with which you can create, delete, and manage user accounts and groups directly from the terminal. An overview of the important Linux commands for user account management is put together for you here. You’ll also find Linux terminal commands in this category that enable you to access code with other user rights, including the super-user root.




Create user account

The simplest option for creating a user account is provided by the command line program adduser. This requires root permissions, and is used according to the following syntax:


Use adduser without options to automatically create a user ID, home directory, and user group with the same name, in addition to the new user account.


adduser test

Terminal output:

Adding user 'test' (1001) …

Adding new group 'test' (1001) …

Adding new user 'test' (1001) with group 'test' …

Creating home directory '/home/test' …

Copying files from '/etc/skel' …

This is followed by an interactive dialog in which you can define the password and other user information (real name, office number, telephone number, etc.).

This automation can be customized or prevented through additional options.

The perl script adduser is based on the low-level program useradd, and offers the same functions in a user-friendly form.


Customize additional user information

The command line directive chfn (short for change finger) allows you to customize additional information on a user account, such as the real name, office number, and private or work telephone numbers.

The general syntax of chfn reads:


The command has to be executed with root permissions.

Which user information will receive a new value is defined with the help of the option -f (real name), -r (office number), -w (work phone), and –h (private phone).

In the following example, the old office number of peter23 is overwritten with the value 122.

chfn -r “122” peter23


Change standard shell

The command line directive chsh (short for change shell) changes the login shell of a chosen user.

The syntax reads as follows:


To customize the login shell of a user, use chsh with the option -s. This directs the path to the desired shell (i.e. /usr/bin/fish).

Notice: Users without root permissions can’t change their own shell. If you would like to change the shell of another user, execute the command with root permissions.


sudo chsh -s /usr/bin/fish peter23

For the user peter23 the shell fish (friendly interactive shell) is defined as default.

Changing the shell doesn’t take effect until the user logs out and back in again.


Delete user account

The command line program deluser deletes all entries for a selected user account from the system account files.

Calling deluser requires root permissions and uses the following syntax:



deluser peter23

The user account peter23 is now deleted.

If you would also like to delete all files from the home directory of the user, then use the command with the options --remove-home. If you want to delete all user files from the system, use the options --remove-all-files.

deluser --remove-all-files peter23

If you want to pack up the user files before deleting them, use deluser in combination with the option --backup-to and specify the desired directory.

deluser --backup-to /path/to/directory peter23

deluser is a perl script that provides the functions of the low-level program userdel in a more user-friendly form.


Create user group

The command line program groupadd is used to create user groups.

Use groupadd with root permissions according to the following syntax:

sudo groupadd [OPTIONS] GROUPS

Each newly-created group contains its own group ID (GID). Group IDs between 0 and 99 are reserved for system groups. If you want to define the GID for a new user group for yourself, use the command line directive groupadd with the option -g (GID).

In the following example, the group users is created with the GID 1425:

groupadd -g 1425 users

If you want to create a system group, use the option -r (root).


Delete user group

The command line directive delgroup (short for delete group) deletes an existing user group.

The general syntax of delgroup reads:

delgroup [OPTIONS] GROUP

To execute the command, root permissions are required.

The following call will delete the users group:

delgroup users

Similar to deluser, this command is also a perl script that offers the functions of the low-level program groupdel in a more user-friendly form.


Customize user group

Names and group IDs (GID) of existing user groups can be customized via groupmod.

The command line directive is used with root permissions according to the following syntax:


Use groupmod with the option -g to customize the GID. Call the command with the option -n to overwrite the group name.


groupmod -g 1800 users

The GID of the users group is set at 1800.

groupmod -n all users

The users group is renamed all.


Change user groups

The command newgrp (short for new group) allows registered users to change their current group ID without having to log out and back in.

The general syntax of the command reads:

newgroup [-] [GROUP]

If the newgrp command is used with the optional parameter [-], then the group change causes a restart of the user environment – as if the user had logged in again.

Users who use newgrp without group specification change to the default group specified under /etc/passwd.

In principle, a user needs to be a member of the group that they want to change into. Password-protected groups are an exception. If a group is protected by a password, then it’s prompted by the terminal before the change.


Change user password

Use the command line program passwd to change the password of a user or define, check, and change intervals.

The command is based on the following syntax:


If you want to change the password of another user, then you need root permissions.

Use the passwd command without a username to change your own password.


If the password is supposed to be blocked, use the command passwd with the option -l (--lock).

passwd -l USERNAME

Other options give you the opportunity to define a duration for the expiration of passwords (-x) as well as warning (-w) and check intervals (-i).


The following example defines for the user peter24 that the password needs to be reset every 30 days. A warning is issued 5 days before the deadline. If the password isn’t renewed after 30 days, then the password expires and the user account peter24 is blocked after 3 days.

passwd -x 30 -w 5 -i 3 peter24


Run programs with the rights of another user

The command sudo (substitute user do) can set the program call to run with the rights of another user. As a rule, the entry of a password is required for this. The command sudo always asks for the password of the calling user.

If the command is entered without a username, then the superuser root is set as the target user.


Administrators have the option to define who can use sudo and which program calls are allowed in the /etc/sudoers file. A user has to belong to the sudo group to be able to use the sudo command.

If you want to select a different target user, then use sudo with the option -u and the desired username.


Such a user change is only possibly if it’s allowed in /etc/sudoers.

If you want to permanently change the command to run with administrator rights in the root shell, use sudo with the option -i.

sudo -i

The sudo command is useful because it allows users to execute previously defined commands as root users without having to enter the root password.


Work with the rights of another user

The command su also allows for a temporary user change to run a program call with the rights of a target user. As opposed to sudo, the command is not directly executed. Instead, a change of identity occurs. Instead of asking for the password of the calling user, the target user password is requested. To call programs as a superuser, a user needs the root password of the system. Also unlike sudo, su can’t be restricted to a set of pre-defined program calls set by the administrator.

The general syntax of the command reads:


A call without a USERNAME selects root as the target user.


Customize user account

The command line directive usermod gives you the option to edit previously created user accounts.

Use usermod with root permissions according to the following syntax:


Which modifications are intended can be defined with the help of options.

Change username (-l NEW_NAME):

usermod -l peter24 peter23

Create a new home directory (-d DIRECTORY) and move the old files (-m):

usermod -d /path/to/directory/peter24 -m peter24

All files from the old home directory will be moved to the new home directory.

Block user (-L):

usermod -L peter24

The password of the user peter24 is blocked.

Include users in groups (-a) and maintain all other group memberships (-G):

usermod -aG users peter24

Peter is added to the users group.

System commands

In the system commands category, you’ll find the basic Linux commands for system control. Use the following commands to restart and shut down the system from the terminal – and control them with a timer, if desired.


Most command line directives for system control must be run with root permissions.




Create log entries

With the command line program logger, entries in the system log can be created.

Use logger according to the following pattern:


Find the system log under /var/log/syslog.


Restart the system

The command line directive reboot causes a restart of the system. To trigger a restart, the command has be executed with root permissions.

reboot [OPTIONS]


Automatically start and shut down the system

The command line directive rtcwake allows you to start and shut down the system according to a timer.

The command is based on the following syntax:

rtcwake [OPTIONS] [MODE] [Time]

Choose a particular mode (-m MODE) for the system to move to at a particular time in seconds (-s TIME IN SECONDS). You also have the option to wake up your system at a precisely defined time (-t UNIXTIME).

Example 1:

rtcwake -m standby -s 300

The system will be put in standby mode for 5 minutes (300 seconds).

Example 2:

rtcwake -m off -t 1490997660

The system will be shut down and ‘woken up’ at the Unix time 1490997660. This corresponds to the following date: 4/1/2017 - 12:01:00 A.M. The Unix time is the number of seconds since 1/1/1940 at 12:00 A.M. Because information is hard to find in Unix, it’s recommended to translate to the date command (listed below).

rtcwake -m off -t $(date -d '20170401 00:01' +%s)


Shut down the system

The command shutdown can be used by the root user to shut down the system.

The command is based on the following syntax:


If you want to induce a shutdown, you have the option to define a time that the system should be turned off. For this, use either a concrete time input (hh:mm) or a countdown (+m).

Other users on the system will get a shutdown message. This can be accompanied by a personal message, if needed.

In the following example, the system will be shut down in 10 minutes:

shutdown +10

If the command shutdown is used with the option -r, the shutdown of the system is followed by a reboot.

shutdown -r +10

System information

In the system information category, we’ve collected command line programs with which you can obtain information and status reports, giving you a comprehensive overview of the state of your system.




Retrieve system time

The command date outputs the system time including the date.


If you want to work with a particular time in the context of a program call (see rtcwake), define this with the help of the option -d 'DATE'. In addition, various options are supported that can transfer date and time data to a desired format.

For example, use the option +%s to output a date in Unix time (number of seconds since 1/1/1970 00:00:00 UTC).


date -d '20170427 11:29' +%s



1493285340 Unit time corresponds to 4/27/2017 - 11:29:00 A.M.


Retrieve free hard drive space

Use the command df (disk free) according to the following pattern to display the free hard drive space of the attached partitions.


If the command is used in combination with a particular file, the system only specifies the free space on the partition where the file is located.


The option -l (local) restricts df to the local file system. It also supports options that let you customize the output format. For a readable output, it’s recommended to use the option -h (human readable): i.e. 3K 124M 1G.


Retrieve messages from the core circular buffer

The program dmesg (short for display message) outputs core circular buffer messages in the terminal and allows you to localize hardware and driver failures.

Use dmesg according to the following pattern:

dmesg [OPTIONS]

The dmesg output contains all messages of the boot routine, and is accordingly long. The command line program is often used in combination with a pager, like more, less, or tail.


dmesg | tail

The dmesg output is delivered to the pager tail with the help of the pipe operator (|). This leads to only the last 10 messages in the terminal being output.

A combination with the grep command makes it possible to do a targeted search through the messages.


Access occupied hard drive space

If you want to know how much hard drive space is occupied by directories on your system, use the command du (short for disk usage) according to the following pattern:


Specifying a particular directory is optional. The occupied hard drive space is output with the option -h for a human-readable format.


Retrieve memory usage

The command free outputs the memory usage.

The general syntax reads:

free [OPTIONS]

As output, you’ll get two specifications: Mem (Memory) and Swap.

Mem deals with the physical memory of your system. If this is drained, then Linux outsources part of the data saved in the RAM to the hard drive. This is referred to as swap space.

Free also supports the option -h for outputting the memory usage in a human-readable format.


Retrieve host name

Use the command hostname according to the following pattern to display the DNS names of the system.

hostname [OPTIONS]


Retrieve core information

The command line directive uname stands for unix name and is used to access system information from the core.

The command supports various options with which the output can be filtered according to the desired information.

uname [OPTIONS


Retrieve system runtime

If you want to determine how long the system has been running since the last reboot, use the command line directive uptime according to the following pattern:



Access statistics about virtual storage

With the help of the monitoring tool vmstat, you can access information about virtual memory, reading and writing procedures on the disc, and CPU activity.

Call vmstat according to the following syntax to output the average values since the last system start.

vmstat [OPTIONS]

vmstat also offers a continuous monitoring mode that accesses system values as often as requested in a desired time interval in seconds.

vmstat [Options] [INTERVAL [REPETITIONS]]


vmstat 4 8

The request takes place in eight passages every four seconds.

If you want to stop the continuous request, use the key combination [CTRL] + [C].

Hardware information

Linux commands in this category deliver detailed information about the hardware components that form the foundation of your system.




Output processor information

Use lscpu (short for list cpu) according to the following pattern to output information about the CPU architecture in the terminal.

lscpu [OPTIONS]

For possible options, refer to your operating system’s manual.


Output hardware information

The command lshw stands for list hardware and outputs information about the hardware components in the terminal. The information includes CPU, memory modules, and devices such as sound cards, graphics cards, or drives that are connected to PCI, USB, or IDE interfaces.

Use lshw according to the following syntax:

lshw [OPTIONS]

The command supports various options for customizing the output format (-html, -xml, -short, -businfo) as well as the range of information (i.e. –sanitize to hide sensitive information).


Output information on PCI devices

Use lspci (short for list pci) according to the following pattern to output detailed information about PCI devices.

lspci [OPTIONS]

For possible options, refer to your operating system’s manual.


Output information on USB devices

Use lsusb (short for list usb) to output detailed information about USB devices in the terminal.

lsusb [OPTIONS]

For possible options, refer to your operating system’s manual.

Process management

On Linux, the instance of a running program is called a process. The following terminal commands are part of the standard repertoire of the process management, and allow you to supervise all processes on your system easily from the terminal and control as necessary.




Request and customize real-time attributes

The command line program chrt deals with continuous process controls and makes it possible to identify and customize the real-time attributes (scheduling regulation and priority) of running processes, or execute commands and their arguments with specified real-time attributes.

The general syntax of the command reads:


Use chrt without entering a priority and with the option –p to identify the real-time attributes of chosen processes:

chrt -p PID


chrt -p 1234

chrt outputs the real-time attribute of the process 1234.

The command is usually used with the following pattern to execute a command and its arguments with a particular real-time priority:



chrt 99 firefox

The program Firefox is started with a real-time priority of 99.

If, on the other hand, the real-time priority of a process that’s already running is supposed to be adapted, then use the following syntax:



chrt -p 20 1234

The real-time priority of the process 1234 is set at 20.

chrt also offers the possibility to set or define the scheduling regulation of running or newly started processes with the help of options.

chrt uses SCHED_RR (Round Robin, explicitly with the option -r) as the standard value of the scheduling regulation. This means that all computation-ready processes successively get an allocated CPU time for a certain interval. This is known as a time slice, and specifies how long a process can run until it’s replaced by another process. The size of the time slice for a process depends on its priority. Linux offers 140 priority levels for processes (0 = highest priority, 139 = lowest priority). The priority levels 1 to 99 are reserved for processes with real-time priority. User processes are usually carried out with a priority level from 100 to 135. This corresponds to a nice value of -20 to +19 (see nice).

In addition to SCHED_RR, Linux also has SCHED_FIFO (option -f) for another scheduling regulation for real-time processes. Like SCHED_RR, SCHED_FIFO works as a first-in/first-out algorithm. This way doesn’t use time slices. Processes started with SCHED_FIFO run long enough that they either finish or are displaced by a process with a higher real-time priority. Displaced processes return to the end of the queue.


Assign I/O scheduling classes

The command line directive ionice is used to influence the priority of a process that uses the I/O interface of the core.

The general syntax of the command reads:


To be able to execute ionice, you need root permissions.

The command distinguishes between three scheduling classes that are passed on using the -cZAHL option. Possible values are 1, 2, and 3.

1 = Real time: The I/O action is executed immediately.

2 = Best effort: The I/O action is executed as quickly as possible.

3 = Idle: The I/O action is only executed when no other process is taking I/O time.

The PID of a running process is passed on with the option -pPID.


ionice -c2 -p1234

The scheduling class 2 (best effort) is passed to the process with the PID 1234.


Stop and finish process via PID

kill is a command line program with which processes can be stopped and finished.

The command is passed on according to the following pattern with a desired signal and the ID of the chosen process.


Common signals are:

TERM: Causes a process to end itself (Standard)

KILL: Forces the end of a process (through the system)

STOP: Stops a process

CONT: Allows a stopped process to continue

The following call sends a signal to the process 1234 that prompts it to end itself. Since no signal is given, kill sends the standard signal TERM.

kill 1234

Always give processes the chance to end themselves, and only force the action via KILL if the affected process doesn’t react as intended.

kill -KILL 1234

If you only want to stop 1234 for a little while, use the following call to pause or restart the process:

kill -STOP 1234

kill -CONT 1234

Use the command kill with the option -l (--list) to show all possible signals that can be given to processes via kill.


Stop and finish processes by name

Use killall in combination with a particular search term to only end the processes whose names coincide (the first 15 characters are used to match).


The option -e (--exact) allows you to extend the match to all characters of the process name.


Define process priorities

The command line directive nice indicates a process value between -20 and +19 at the start of a process in integer steps, after which the available computing power of the system is distributed. The range of -20 to +19 corresponds to the Linux priority levels 100 to 139. A process with a nice value of -20 has a higher priority than a process with a nice value of 19.

The solo syntax reads:


Without additional specification, every process starts with a nice value of 0- Use the option -n to define the process priority. It should be noted that negative priorities can only be assigned with root permissions.

In the following example, the editor nano is started with a priority of 4:

nice -n 4 nano


Delete process from session

Normally, all of a user’s dependent processes are automatically ended as soon as the terminal session is closed (i.e. via exit).

The command line directive nohup (short for no hangup) deletes a command from the current session and allows you to keep it running even when you log out of the system. The associated HUP signal (hangup) that normally causes a process to automatically terminate is suppressed to nohup.

The program call is based on the following pattern:



Identify PID via search term

The command line program pgrep matches the list of running processes with a search term and outputs the respective PIDs if there are matches.

The general syntax reads:

pgrep [OPTIONS] Searchterm

By default, pgrep outputs the PIDs of all processes that contain the search term.


pgrep ssh

This will list all the processes that contain the search term ssh in the process name.

If the search is to be limited to only exact matches, then use the command along with the option -x.


pgrep -x sshd

This only lists the processes that are called exactly sshd.

If you would like to obtain the PID in addition to the process name, use pgrep with the option -l.

Similarly to grep, pgrep supports search terms based on regular expressions.


Identify PIDs

The command line program pidof outputs the process identification numbers (PIDs) of all of a program’s processes.

Identify PIDs via pidof according to the following pattern:


With the following call, the IDs of all running processes of the program nano are output in the terminal.

pidof nano

If you would like to output only the first process ID, use pidof in combination with the option -s (short for single shot).


Stop and finish processes via search term

Like kill, the command pkill also sends a signal to a chosen process. The addressing isn’t done by PID, though. Instead, a search term is given that matches the name of the running process. This can also be formulated as a regular expression.

pkill forwards the standard signal TERM, as long as no other signals are defined. The general syntax of the command reads:


Additional options can be used to limit the command to the processes of a particular user (-U UID), the sub-processes of a particular parent process (-P PID), or the newest (-n) or oldest (-o) processes.

While pkill addresses all processes whose names contain the search term, the killall command only targets processes that are an exact match.


Obtain list of all running processes

The command ps outputs a list of all running processes in the terminal.


If you need a detailed output, use ps with the options -f (detailed) or -F (very detailed).

For additional options, refer to your operating system’s manual.


Obtain running processes as a tree structure

Use pstree to display all running processes in a tree structure.

The general syntax of the command reads:

pstree [OPTIONS]

The format and range of the output can be customized using various options.


Customize priorities of running processes

The command line directive renice allows you to customize the priority of a running process.

The general syntax reads:


The addressing takes place with the help of options concerning the process ID (-p PID), group ID (-g GID), or a username (-u USER).


renice 12 -p 1234

The process with the ID 1234 is assigned the priority 12.

renice 3 -g 3456

All running processes of the group with the GID 3456 are assigned a priority of 3.

sudo renice -6 -u peter24

All running processes of the user peter24 are assigned a priority of -6.

If renice is used without options, the default value -p is assumed and the following string interpreted as a process ID.


Delay process execution

The command line directive sleep allows you to disrupt the current terminal session for a given time.

The general syntax of the command reads:


If you use sleep without a suffix, the given number will be interpreted as time in seconds (s). You also have the option to disrupt the terminal session for minutes (m), hours (h), or days (d).

The following call disrupts the session for 4 minutes:

sleep 4m

The command is useful, for example, for delaying the execution of a subsequent command:

sleep 1h && reboot

The system waits one hour, and then afterward carries out the command reboot, which causes the system to restart.


Assign processes to certain processors

The command line directive taskset is used for advanced process control, which is used in multiprocessor systems to assign processes or commands to specific processors.

The command requires root permissions and uses one of the following patterns:


taskset [OPTIONS] -p PID

Assigning a process or command to a processor happens using a hexadecimal bitmask. For example:

0x00000001 = Processor #0

0x00000003 = Processor #0 and #1

0xFFFFFFFF = All processors (#0 to #31)

Since assigning via bitmask like this isn’t very intuitive, taskset is generally used with the option -c (--cpu-list) to enable a numerical assignment of processors (i.e. 0, 5 7, 9-11).

The following command tells the process 1234 to use processors 1 and 2:

taskset -p 1234 -c 1,2


Dynamic process overview

The command top calls a dynamic overview of all running processes.

The call is based on the following pattern:


The output of the process information can be adjusted using various options. The top process overview (among others) supports the following hotkeys to sort through the outputs:

[P] = Sorts the output according to CPU load

[M] = Sorts the output according to storage requirements

[N] = Sorts the output numerically by PID

[A] = Sorts the output by age

[T] = Sorts the output by time

[U USERNAME or UID] = Filters the output by respective user

Use the hotkey [H] to display a help page, or [Q] to close the process overview.


Do you want to use your overview to keep track of multi-page file content? With a command line program from the pager category, you can select which sections are displayed in the terminal and scroll through the file in interactive mode if necessary.  




Output the first lines of a file

The pager head is used to output the first part of a file.

The general syntax of the command reads:

head [OPTIONS] File

Use the option -n NUMBER_LINES to define how many lines are to be output, starting at the beginning.


head -n 3 example.txt

This will output the first three lines of the file example.txt. Without line specification, head outputs the first 10 lines of the given file.


Display text files in the terminal

The command line program less enables the display of the content of a text file in the terminal.

The general syntax reads:


The output is automatically in interactive mode. This allows you to scroll through the selected document or search by keyword.

The [Q] key ends the interactive reading mode. Other control keys and available options can be found in the program’s manual.


Display text files in the terminal

The pager more fulfills the same function as less, but offers a smaller range of functions.

Use more according to the following pattern to call a text file and its content in the terminal:


The command line program always displays a complete screen page of the chosen file. If a file contains multiple pages, more starts an interactive mode that allows you to scroll through the document using control keys or search by keyword.

The [Q] key ends the interactive mode. Other control keys as well as available options can be found in the manual of your operating system.


Output the last lines of a file

While head displays the first 10 lines of a chosen file by default, tail outputs the last 10.

Both pages are used according to the same pattern (refer to head).


Under Linux, you don’t need a graphical text editing program to customize configuration files, edit code snippets, or draft short notes. Simple text editors can be easily called up in the terminal without time delays. Here we present three programs that you should know.




The text editor Emacs

Emacs is a cross-platform text editor, which can be expanded as desired by a programming interface.

By default, Emacs starts with a graphical user interface, but can also be opened in the terminal using the option –no-window-system.

emacs –no-window-system

Emacs has an available integrated tutorial that you can call with the key combination [CTRL] + [H], [T].


The text editor Nano

Nano is a GNU reproduction of the terminal-based text editor Pico, used in the context of the mail client Pine. Nano offers a smaller range of functions than comparable editors (i.e. Vim), but is characterized by a particularly user-friendly handling.

The general syntax of the program call reads:


The program opens the given file in an editing window in the terminal.

If you call Nano without file names, a new text file can be created that’s stored in the currently selected directory.

nano [OPTIONS]

The key combinations for controlling the program are listed at the bottom of the editing window. Other information about Nano can be found in the program’s manual.


The text editor Vim

Vim (short for Vi Improved) is a further development of the text editor Vi that stands out due to numerous extensions such as syntax highlighting, a comprehensive help system, native scripting, automatic code completion, and visual text selection.

The open source program offers various modes of operation for editing pure text files, and can be used either in the terminal or as a stand-alone application with a graphical user interface (GVim). A central application area of the program is the editing of program code.

If you start Vim in the console, the operation is done via keyboard. Generally, the program is called together with a text file according to the following pattern:


Opened files load Vim into a buffer. All changes that you make to the open file are also kept here. If you open Vim without specifying a file, the program starts with an empty buffer. The original file is not adapted until the memory operation is provided by the corresponding key combination. If no files exist that match the name given in the program call, one is newly created as part of the memory process.

Vim offers the program vimtutor as a comprehensive introduction, which is also started from the command line.

Our basics article on the text editor Vim also offers additional information on the installation and various operating modes of the program.

Network management

Network management is also managed easily from the terminal in Linux. Whether you want to test the connection, request DNS information, configure the interface, or transfer files to another computer in the network, with the following programs a single command is sufficient to put your project into motion.




Display and manipulate the ARP cache

The command line program arp allows you to access and manipulate the ARP cache of your operating system.

Use arp without a modifier to output the content of the ARP table in the terminal.


You can also use the option -a to limit the output to entries for a particular hostname (or an IP address).



arp -a

If you want to create an ARP entry, use a program call with the option -s according to the following pattern:



arp -s 00:05:23:73:e6:cf

If a particular entry is to be deleted, use arp with the option -d:



Request DNS information

dig is a lookup tool that can be used to request information from the DNS server and output it in the terminal.

The command line program is generally used according to the following syntax to request the IP address and other DNS information on a given domain name:


SERVER is the DNS server that is to be searched for the desired information. If no server is given, dig identifies the standard DNS server from the file /etc/resolv.conf.

DOMAIN stands for the domain name from which the DNS information should be identified.

TYPE is used to specify the type of query, i.e. ANY (all entries), A (IPv4 record of a host), or AAAA (IPv6 record of a host). The standard request type is defined as A.

Use dig with the option -x to request the domain name for a given IP address as part of a reverse lookup.


The arguments NAME, TYPE, and CLASS are not needed in this example.


Transfer files via FTP

With the command line program ftp, most Linux distributions have a pre-installed client program for data transfer via FTP (File Transfer Protocol). This gives you the possibility to exchange files between the local system and another computer in the network.

Use ftp according to the following syntax to establish a connection to the FTP server of the target computer:


The addressing takes place via host name or IP address. Specifying a port number is optional.

In general, you’re asked for a username and corresponding password when establishing a connection.

If the login is successful, ftp starts a command line interpreter that accepts user input in the form of commands. The program supports various commands for searching and managing the file system of the target computer, as well as transferring files from one system to another.


Obtain and configure network interfaces

The command line program ip is part of the program collection iproute2, with which network interfaces are requested and configured via the terminal.

The general syntax of the command reads:


Which action is carried out by ip is defined with the help of objects, subcommands, and their arguments.

The program supports various objects, such as address (IP address), link (network interface), route (entry in the routing table), or tunnel, to which subcommands such as add, change, del, list, or show can be added.

For example, if you would like to access the IP address of a particular network interface (i.e. eth0), use the command ip in combination with the object address, the command show, and the argument dev eth0:

ip address show dev eth0

You can also give the objects and commands in shorthand:

ip a s dev eth0

If you would like to output all information to a network interface (i.e. eth0), use the command line directive ip with the object link, the command show, and the argument dev eth0:

ip link show dev eth0


ip l s dev eth0

To activate or deactivate an interface like eth0, enter the following:

ip link set eth0 up

ip link set eth0 down

With its large range of functions, the program collection iproute2 replaces a number of older network tools such as ifconfig, route, and netstat.

A list of all possible options, objects, subcommands, and arguments for the command line directive ip as well as information on other iproute2 programs can be found in the manual of your operating system.


Obtain and configure WLAN interfaces

The command line program iw is used for the configuration of WLAN interfaces and is established as a current alternative to iwconfig.

The call is based on similar syntax to that of the ip command:


Possible objects are:

dev NAME_OF_INTERFACE = Network interface

phy NAME_OF_DEVICE = WLAN device (by name)

phy#INDEX_OF_DEVICE = WLAN device (by index)

reg = Regulatory agent for the configuration of regional and country settings

Use iw with the command help to display the program syntax as well as possible options and commands.

iw help

The following is an application example of the command line program iw:

Output device settings of all WLAN interfaces:

iw list

Access connection status (transfer rate and signal strength) of a WLAN interface (i.e. wlan0):

iw dev wlan0 link

Scan WLAN environment:

iw dev wlan0 scan

Use iw in combination with the command scan to output all WLAN networks in the reception area as well as their properties (radio channel, encryption, signal strength, etc.).

Read out regional settings:

iw reg get

Modify regional settings:

iw reg set US

Request device properties (i.e. from wlan0):

iw list dev wlan0

Detailed device properties:

iw dev wlan0 station dump

Query events:

iw event

The options -f, -t, and -r deliver extended output with error messages regarding connection status and time.


Obtain network interface status

The command line program netstat is used to query the status of network interfaces.

The general syntax of the command reads:

netstat [OPTIONS]

Use netstat without option to output all open sockets in the terminal.

You can also use the following options to see the routing table (-r), interface statistics (-i), masked connections (-M), or network link messages (-N).

An alternative to netstat is contained in the iproute2 program collection program ss.


Obtain DNS information

Like dig, nslookup is also a name resolution service. The command line program is available in two modes: interactive and non-interactive.

The interactive mode starts when the command nslookup is entered in the terminal without any additional information.


The program now accepts commands. For example, type a hostname (domain) to get the corresponding IP address.

You can also start a reverse lookup request by entering an IP address and then outputting the associated hostname.

The program nslookup automatically uses the DNS server pre-installed in the system.

Enter the command exit to end nslookup.

If you want to use nslookup in non-interactive mode, call the program in combination with a hostname or IP address.

nslookup [OPTIONS] [HOST/IP]

Since the program is officially outdated, users are encouraged to rely on dig instead.


Test network connection

Use the command line program ping to test the accessibility of other computers in the network.

The command is based on the following syntax:


To check the network connection, ping sends a small data package to the given target system (hostname or IP) and analyzes the time until the answer is received.

Together with the round-trip time (RTT) – the time span between sending the data package and receiving an answer – ping also writes the IP address of the target system in the terminal. So the command line program is also suitable for determining the IP address for a domain.

If ping is used without options, then the program runs until it’s ended manually with the key combination [CTRL] + [C], and sends the target system a ping request every second.

If you want to define an end time when you call the program, use the options -c NUMBER (number of ping requests that will be sent) or -w SECONDS (time span in seconds, after which ping will end itself).


Display and edit the IP routing table

With the command line program route, the IP routing table of the core can be requested and edited.

The command is based on the following syntax:

route [OPTIONS]

route [OPTIONS] [add|del] [-net|-host] TARGET

Use the command without options to display the complete routing table of the core:


If you want to set a route to a network, use the subcommand add.

route add -net

If the target is a subnet, then the subnet mask has to be specified using the netmask MASK option:

route add -net netmask

You can also set up a route to a computer:

route add -host

If the system is available over multiple network interfaces, then the option dev INTERFACE needs to be used to specify which interface should be used:

route add -net netmask dev eth0

If a target can only be reached via a router, this must also be specified with the gw ROUTER option.

route add -net netmask gw

If you want to delete a route, use the subcommand del.

route del -host


Synchronize files

The command line program rsync enables you to synchronize files locally or over a network. For this purpose, the size and modification time of the concerned files are compared.

If the source and the target are located on the same system, then different files are copied completely. For synchronization over a network, rsync uses a delta transfer algorithm so that only altered file components have to be transferred from the source file carrier to the target system.

The syntax of the call reads:



rsync -a home/user/documents/ /home/user/backup

All files from home/user/documents are compared with the files in the directory /home/user/backup.

The command rsync is generally performed with the option -a, which ensures that all subdirectories and symbolic links are copied and all user rights take effect.


Transfer files via SFTP

The command line program sftp functions like ftp to transfer data in the network. But here, all operations are performed via an encrypted SSH connection (secure shell).

Like ftp, sftp creates a connection to a target computer in the network and then starts an interactive command mode.


Transfer files via SCP

With scp (short for secure copy), another program for secure data transfer in the network is available directly via the terminal: scp copies data from one computer to another and uses the network protocol SSH.

The client program functions in the same way as the file option cp, but is used system-wide according to the following syntax:

scp [OPTIONS] FILE [[user@]remote_host:]PATH

When specifying the path of the remote computer, the username and the respective hostname are placed in front. Local files are explicitly addressed using relative or absolute paths.



The file image.jpg is copied from the local images directory to the archive directory on the target computer with the address

The program scp also supports data transfer in the opposite direction as well as between two remote systems.

scp [OPTIONS] [[user@]host:]FILE PATH

scp [OPTIONS] [[user@]host1:]FILE [[user@]host2:]PATH

Additional options allow you to make adjustments to the transfer mode and the encryption settings.


Track data packages

Use the command line directive traceroute according to the following pattern to track the transport path of an IP data package between your system and a destination computer.


Via traceroute you can identify which router and internet nodes an IP package passes on its way to the target computer – for example, to investigate the cause of a delay.


Output terminal names

The command line directive tty outputs the file names of the terminal that are defined as the standard input.

The general syntax of the command reads:


Archive and compress

Linux offers various technologies with which files can be packed and compressed in archives. It should be noted that not every archive contains a compression. So tar – a program for the archiving of files – is usually combined with a compression program like gzip, bzip2, or xz.




Write and extract files in the tar archive

The command tar stands for tape archiver, a program that was originally developed to secure data on tape drives. Even today, tar is one of the most popular programs for archiving data under Linux.

The program allows you to write various files and directories sequentially into a tar file and use it as a backup for recovery if needed. Unlike the zip format common in Windows, all user rights of the archived file are retained even after unpacking.

The command line program tar is called according to the following syntax:


If you want to create a new archive, use tar with the options -c (create new archive) and -f (write archive to a given file or read from it).

In the following example, the files file1.txt and file2.txt are written into the newly created archive example.tar.

tar -cf example.tar file1.txt file2.txt

If you want to see the content of an archive, use tar with the options -t (display archive content), -v (detailed output), and -f (see above).

tar -tvf example.tar

If archived files are to be unzipped into the current folder, then use tar with the options -x (extract files from archive) and -f (see above).

tar -xf example.tar

tar offers more possibilities with -j (bzip2), -J (xz), -z (gzip), and -Z (compress) that allow you to compress or decompress archives when you call another program during the packing and unpacking processes.

In the following example: the files file1.txt and file2.txt are archived in example.tar.gz and compressed with gzip.

tar -czf example.tar.gz file_1.txt file_2.txt

The following command extracts and decompresses all archived files in example.tar.gz.

tar -xzf example.tar.gz

gzip / gunzip

Compress or decompress files with gzip

gzip (short for GNU zip) is a program with which you can easily compress or decompress files via the command line.

The general syntax of the command reads:


For example, use gzip according to the following pattern to transfer the file example.txt to the compressed format example.txt.gz:

gzip example.txt

Note that by default, gzip deletes the original file as part of the packing process. Prevent this by using the option -k.

gzip -k example.txt

The program can be used for multiple files at the same time, if necessary. Each output file is converted into a separate gz file.

So the command…:

gzip example_1.txt example_2.txt example_3.txt

…generates the files example_1.txt.gz, example_2.txt.gz, and example_3.txt.gz. 

bzip2 / bunzip2

Compress and decompress files with bzip

A popular alternative to gzip is the command line program bzip2. This uses the same syntax as gzip, but is based on a three-stage compression process which allows for a significantly higher compression ratio.

First, the given files are wrapped block-wise by the Burrows-Wheeler transformation and then by the move-to-front transformation. The actual data compression finally happens with a Huffman coding.

Files that are compressed with bzip2 use the file ending -bz2. Use bzip according to the following pattern to compress files:


bzip2 can also be applied to tar archives.

The decompression is analog with gzip and runs with the help of option -d. The command bunzip2 is also available.

Users pay for the high compression ratio with a comparatively long runtime.


Compress and decompress files with xz

The command line program xz converts files in the same-named data compression format xz. The program call uses the same pattern as gzip and bzip2.


Files that are compressed with xz use the file ending .xz. The decompression functions as with gzip and bzip with the option -d. The command unxz can also be used.

Like gz and bz2 files, xz files are also not archive files. If you would like to write multiple files into the same compressed xz file, you also have to rely on the archiving tool tar with this compression program.

xz supports various compression algorithms. The Lempel-Ziv-Markov algorithm (LZMA/LZMA2) is used by default.


Write and extract files to archive file

The archiving program cpio (short for copy in, copy out) allows you to write data in an archive file (.cpio) and extract data from it.

A detailed description of the command line programs listed here can be found in our basics article on the topic “Archiving and Compressing Using Linux”. Additional information on compression methods, as well as the definition of deduplication, can be found in our article on data reduction.

Partition management

If you want to access a file system on another partition in Linux, you first have to integrate it into the directory structure of your operating system. This is called “mounting” a partition. If necessary, this can happen via the graphical user interface. Command line programs like lsblk, blkid, and mount also offer the ability to request information about connected block storage devices and to mount or unmount them when necessary.



mount /unmount

Integrate file systems

If a file system is to be integrated in the directory structure of the operating system via the directory structure, then the command line program mount is used on Linux.

The general syntax of the command reads:


DEVICE = Path to the device file of the storage device that you want to mount as the partition.

MOUNTPOINT = The location in the directory structure of your operating system where you want to mount the partition. The mountpoint is usually specified as an absolute path.


mount /dev/sdd /media/usb

The device sdd is mounted in the directory /media/usb.

In general, Linux automatically recognizes the respective file system of the device. If this isn’t the case, then the option -t gives you the option to explicitly share the file system (i.e. ext4):

mount -t ext4 /dev/sdd /media/usb

If a previously integrated file system is to be unmounted, then use the command unmount:




If you want to output all file systems that are integrated in your operating system, use the command mount with the option -l.

mount -l

The output can be limited to file systems of a particular type via -t.      


List information on connected block storage devices

Use the command lsblk (short for list block devices) to represent all connected block storage devices and partitions as a tree structure. These don’t necessarily have to be involved.

The call is based on the following syntax:

lsblk [OPTIONS]

The output includes the following information:

NAME = Device name (i.e. sda) or partition name (i.e. sda1, sda2, etc.)

MAJ:MIN = major:minor Device number

RM = Exchange medium (1 = applicable, 0 = not applicable)

SIZE = Device storage size

RO = Read-only device (1 = applicable, 0 = not applicable)

TYPE = Device type

MOUNTPOINT = Mounting point

If necessary, the output and a list of desired attributes can be individually modified using the -o (--output) option to retrieve additional information, like the identification number (UUID), file system (FSTYPE), or the state (STATE).



In the standard settings, empty storage devices are bypassed. If you also want to include these in the overview, use lsblk in combination with the option -a (--all).

If you only want to request information on a particular device, use lsblk according to the following pattern:



lsblk /dev/sda


List information on connected block storage devices

Similar to lsblk, blkid also outputs information on connected block storage devices.

Use blkid according to the following pattern to obtain the identification number (UUID) and file system type (TYPE) of all connected block storage devices.

blkid [OPTIONS]

For tabular output, use the -o option in combination with the value list.

blkid -o list

You can also limit blkid to a chosen device:



blkid /dev/sda1


Bit-exactly copy files, partitions, or volumes

The command line program dd enables a copying process in which data is read out bit for bit from an input file (if) and written into an output file (of).

The program call is based on the following syntax:

dd if=Source of=Target [OPTIONS]

As the source and target, you can specify individual files as well as entire partitions (i.e. /dev/sda1) or a complete storage device (i.e. /dev/sda).


dd if=/dev/sda5 of=/dev/sdb1

The complete fifth partition of /dev/sda is copied bit-exactly from the first partition of /des/sdb.

The copying process can be limited to any number of memory blocks of the desired size via options.


The following list contains additional Linux basic commands that don’t belong to any of the previous categories.




Define nicknames for program calls

The command line program alias enables you to define nicknames for program calls.

Use alias according to the following pattern:


Replace the placeholder COMMAND with any command line directive, including options. This will link the inserted string for the placeholder NICKNAME.


alias ll='ls -l'

The string ll is defined as the alias for the command ls with the option -l (detailed output).


Run time-controlled command

Call the command line program at according to the following pattern to run a time-controlled command.



at 10:00 AM 6/22/2017

Then enter the command and close the interactive mode with [CTRL] + [D].


Display calendar

Use cal according to the following pattern to output a calendar in the terminal.

cal [OPTIONS] [[MONTH] Year]


cal 12 2017

The system outputs a month overview for December 2017.


Output string to the standard specification

Use the command line directive echo to output strings line-by-line on the standard output (usually the terminal).

The general syntax of the command reads:



Prepare text files for printing

Use the command line program pr to prepare text files for printing.

The general syntax of the command reads:

pr [OPTIONS] File

In the standard settings, pr generates a page header that contains the file name, current date, and page number.


Record terminal session

The command line program script allows you to record a terminal session in the file typescript. If there’s already a recording of a previous session in typescript, then it’s overwritten.

The recording automatically starts with the program call:


Use the key combination [CTRL] + [D] to end the recording.

If you would like to save the recording in another file instead of in typescript, call script in combination with a file name or path.

script FILE


Output numerical series

Use the command seq to output a numerical series in the standard output. Define a start value, an end value, and an increment (optional).



seq 0 2 100

The program counts from the start value 2 to the end value of 100 in increments of 2.


Installation help for standard applications

The command line program tasksel serves as installation help for standard applications (mail server, DNS server, OpenSSH server, LAMP server, etc.). Use the tool to automatically install all packages and programs required for a task in the correct order.

Call tasksel with the option --list-tasks to output a list of all available standard applications.

tasksel --list-tasks

If you want to access more information about a standard application on the list, use tasksel with the option --task-desc and the corresponding task.


tasksel --task-desc mail-server

This outputs information about the “mail-server” task.

If you want to list all packages that belong to the “mail-server” task, use tasksel in combination with the option --task-packages.

tasksel --task-packages mail-server

To install all packages of a standard application, use the subcommand install. This requires root permissions.


tasksel install mail-server

The command line program initiates the installation of all packages that are necessary for the “mail-server” task.


Double program output

The command line program tee is used to double the output of a program. One output is passed to the standard output, and another is written to the file given with the tee command.

The general syntax of the command reads:


tee is usually used in combination with the redirection operator Pipe (|).

ls | tee example.txt

The command ls lists the content of the current directory. The program output is delivered to the command line program tee via Pipe, which displays this in the terminal as well as in the file example.txt.


Measure the runtime of programs

Use the command time according to the following pattern to identify the runtime of programs that you’ve started over the terminal.

time [OPTIONS] Command [ARGUMENTS]


Replace characters in text files

Use tr to delete a desired character set or replace it with another. To do this, tr reads the data stream of the standard input and writes it to the standard output according to the desired modification.

If a character set is to be replaced by another, then tr is used with two arguments.


The second argument (CHARACTERSET2) replaces the first (CHARACTERSET1).

If you want to delete a character sequence, use tr with the option -d and enter the set to be deleted as the argument.


The command line program is usually used in combination with redirection operators (< and >) to make modifications to files.

tr 'a-z' 'A-Z' < example1.txt > example2.txt

tr reads out the content of the example1.txt file, replaces the lower-case letters a through z with upper-case letters, and writes the output in the example2.txt file.


Send messages to all registered users

The command line program wall allows you to send a message to all users registered on a system.

To send a communication, start the program with the following call:


Confirm the program call with [Enter] and enter your message. Then confirm again with [Enter] and send with the key combination [CTRL]+[D].

All users registered on the system receive your message as a broadcast in the terminal.

Note: To be able to receive communications, you have to provide other users with a write access to your terminal. For this, use the command mesg:

mesg [y/n]

Obtain current status:


Provide write access:

mesg y

Deny write access:

mesg n

If you want to send file content to all registered users, use wall in combination with an input redirection and the respective file name:



Periodically run command

The command line program watch allows you to set a command to run at regular intervals.

The program call is based on the following syntax:


The time interval at which the command given in watch will be run is defined with the option -n SECONDS.

End watch with the key combination [CTRL] + [C].

In the following example, the system is instructed to output the workload of the internal memory at 10-second intervals.

watch -n 10 free


Count lines, words, letters, characters, and/or bytes of a text file

The command line program wc short for (word count) outputs the number or lines, words, letters, characters, and/or bytes of a text file, by request.

The overall syntax of the command reads:



wc example.txt


14 18 143 example.txt

If wc is called without options, the output corresponds to the LINES WORDS CHARACTERS FILE. For a filtered output, the command line program supports the options: -l (lines), -c (bytes), -m (characters), -L (length of the longest line), and -w (words).


Convert to standard input in command line

The command line program xargs allows you to transfer the output of a previous command to a new command as an argument. Generally, this is used with the Pipe (|) as a diversion operator.

Use xargs according to the following syntax:


xargs can be used in combination with the command find, for example.

In the following example, find identifies all files in the current directory that fit the search term *.tmp, and outputs their names to the standard output. There, the file names of xargs are accepted and passed as arguments to the command rm.

$ find . -name '*.tmp' | xargs rm

The overview presented here doesn’t claim to be complete, but includes basic Linux commands with selected application examples for everyday work with unix-like operating systems. A comprehensive description of the command line programs presented here, as well as all other commands, can be found in the manual of your operating system. An online version of these help-and-documentation pages are available via the Linux man-pages project from Michael Kerrisk on

We use cookies on our website to provide you with the best possible user experience. By continuing to use our website or services, you agree to their use. More Information.