at (Unix)

In Unix-like computer operating systems, the at command is used to schedule commands to be executed once, at a particular time in the future.

Design

at reads a series of commands from standard input and collects them into one "at-job" which is carried out at a later date. The job inherits the current environment, so that it is executed in the same working directory and with the same environment variables set as when it was scheduled.

It differs from cron, which is used for recurring executions (e.g. once an hour, every Tuesday, January 1 every year). As with cron, many Unix systems allow the administrator to restrict access to the at command.

at can be made to mail a user when done carrying out a scheduled job of theirs, can use more than one job queue, and can read a list of jobs to carry out from a file instead of standard input.

Usage

A sample command to compile a C program at 11:45 a. m. on January 31st would be:

 $ echo "cc -o foo foo.c" | at 1145 jan 31

or

 $ at 1145 jan 31
 at> cc -o foo foo.c
 at> ^D #(press Control-D while at the beginning of a line)

The atq program lists the currently queued jobs, while atrm removes jobs from the queue:

$ atq
1234 2011-08-12 11:45 cc -o foo foo.c user
$ atrm 1234 
$ atq
$

In some Unix-like computer operating systems it uses a daemon, atd, which waits in the background periodically checking the list of jobs to do and executing those at their scheduled time on behalf of at.

The batch command can be used instead of at to only run scheduled jobs if the system's load average is below a certain value.

Windows NT/2000/XP/7 also has an at command (similar to cron), but it is deprecated in favor of Task Scheduler.

See also

• Cron, runs scheduled tasks at regular intervals
• systemd, incorporates atd
• Launchd, Apple Computer's replacement for at
• List of Unix programs

External links

• at: execute commands at a later time – Commands & Utilities Reference, The Single UNIX® Specification, Issue 7 from The Open Group