Unix Tutorial

The following is a quick overview of essential UNIX commands.

File and Directory Permissions Demystified

When doing a long listing (ls –l) of a directory, we see the following entry:

drwxr-xr-x 2 hutchib nsm 512 Apr 20 11:09 bin

This is a directory named bin that is owned by user hutchib and group nsm. Let's explore the first 10 characters in detail.

• The first character is a d, which indicates that we are looking at a directory. Other possibilities for the first character include an - for an ordinary file and an l for a symbolic link (like a shortcut in Windows).
• Characters 2 through 4 correspond to user permissions, or permissions corresponding to the owner of the directory. They are rwx, which indicates that the directory is readable, writable, and executable by the user (the owner of the directory, hutchib).
• Characters 5 through 7 correspond to group permissions. In this example, these permissions apply to the group nsm. They are r-x, indicating that the directory is readable and executable by any user belonging to group nsm.
• Characters 8 through 10 correspond to other permissions. If you are not the owner of the directory, nor a member of the group, the other permissions are applicable. They are r-x, meaning that the directory is readable and executable by everyone else.

What do read, write, and execute permissions mean on a directory?

Read = You are able to list contents of the directory and display contents of files.

Write = You are able to create, modify, or delete files in the directory.

Execute = You are able to change into the directory. (Even if you had read and write privileges
on a directory, if you do not have execute privileges on the directory, you would not be able to get into (cd) the directory.)

Note: If you have write privileges to a directory, you may delete any file in that directory, even if it is not writable, and is owned by somebody else.

What do read, write, and execute permissions mean on a file?

Read = You are able to list the contents of a file.

Write = You are able to modify or delete the file. (If you have permission to write to a file, you
may modify it, but may not delete it, if the directory is not writable.)

Execute = You are able to execute the file. (This is needed if the file is a shell script or Executable.)

How do I change file/directory permissions?

You may change file and directory permissions with the chmod command.

The syntax is:

chmod octal-mode file_or_directory

You may recursively change permissions with the chmod –R command.

When looking at the following table, remember these values when setting file or directory permissions:

Read = 4
Write = 2
Execute = 1

Useful 'vi' Commands for Vim Text Editor

The vi editor has two modes: Input and Command. In Input mode, any characters you type will be entered into the text file. In Command mode, characters you type are used to move around the screen and modify the file. If you are unsure which mode you are in, press Esc, and you will be put into Command mode.

File Extension Primer

You can use the UNIX command file to attempt to determine the file type.

Example:

$ file /home/hutchib/long_cmd.sh
/home/hutchib/long_cmd.sh: executable shell script

$ file /usr/bin/ls
/usr/bin/ls: ELF 32-bit MSB executable SPARC Version 1, dynamically linked, stripped

How do I redirect command output?

Generally, output of your UNIX commands is sent to standard output (STDOUT). You may redirect command output to a file by using:

command > file

If file does not exist, it will be created. If file already exists, it will be overwritten. You could instead append to the end of file by using:

command >> file

In addition to standard output, UNIX has a mechanism called standard error (STDERR). It is sometimes useful to discard standard error when performing commands. For example, when executing a find command on a UNIX system, you normally don't want to see permission denied messages when find attempts to read directories you don't have access to. The following command would redirect standard error to /dev/null, effectively discarding the output:

command 2>/dev/null

If you absolutely do not want the command to generate any output (e.g. a crontab entry), use the following command to send both standard output and standard error to /dev/null:

command >/dev/null 2>&1

How do I put a process in the background?

$ long_cmd.sh
Z[1] + Stopped (SIGTSTP) long_cmd.sh

$ bg %1
[1] long_cmd.sh&

$ jobs
[1] + Running long_cmd.sh

$ fg %1
long_cmd.sh

Execute a shell script called long_cmd.sh, which will take several minutes to complete. Since no interaction with the terminal can occur while this script is executing, press Ctrl-Z to suspend the script. After suspending the script, put it in the background with the bg %1 command. List the background processes with the jobs command. The background process will continue to execute and will complete; the shell will notify you when the job has completed. For this example, you can put the job back in the foreground with the fg %1 command.