The Unix su (subsitute user) command is used to assume the login shell of another user without logging out. It is commonly used to change to root user permissions for administrative work without logging off and back on. Desktop environments such as KDE and GNOME have programs that pop up a password query box before allowing a user to run commands that would typically require such access.

su is commonly run from the command line terminal. When run, su asks for the target user's password, and, if accepted, grants the user access to that account.

A related command called sudo executes a command as another user but does not start a login shell, similar to using su -c.

Great care must be taken by a system administrator to choose a suitable password for the root account, to prevent any possible takeover by a low level user running su. Some Unix-like systems have a wheel group of users, and only allow these users to su to root. This may or may not mitigate these security concerns, since an intruder might first simply break into one of those accounts. GNU su, however, does not support a wheel group; this was done for philosophical reasons. [1]

See also: Unix security

External link

• su(1) manpage