How to use the at command linux

One Time Task Scheduling using at Command in Linux

While working with Linux systems we preferred crontab for scheduling jobs generally. There are another utility at command is very useful for scheduling one time tasks. It reads commands from standard input or script/file which can be executed later once. But we can’t use at command for any recurring tasks. For recurring tasks use Linux crontab.

At command can be useful for shutdown system at the specified time, Taking a one-time backup, sending email as a reminder at the specified time etc. This article will help you to understand the working of at command with useful examples.

Commands used with at:

• at : execute commands at specified time.
• atq : lists the pending jobs of users.
• atrm : delete jobs by their job number.

1. Schedule first job using at command

Below example will schedule “sh backup.sh” command to be executed on next 9:00 AM once.

Use ^d to exit from at prompt.
You can also use the following option to schedule a job. The below command will run “sh backup.sh” at 9:00 in the morning.

2. List the scheduled jobs using atq

When we list jobs by root account using atq , it shows all users jobs in the result. But if we execute it from a non-root account, it will show only that users jobs.

Fields description:
First filed: job id
Second filed: Job execution date
third filed: Job execution time
Last field: User name, under which job is scheduled.

3. Remove scheduled job using atrm

You can remove any at job using atrm with their job id.

4. Check the content of scheduled at job

atq command only shows the list of jobs but if you want to check what script/commands are scheduled with that task, below example will help you.

In the above example, 5 is the job id.

Examples of at Command:

1. Schedule task at coming 10:00 AM.

2. Schedule task at 10:00 AM on coming Sunday.

3. Schedule task at 10:00 AM on coming 25’th July.

4. Schedule task at 10:00 AM on coming 22’nd June 2015.

5. Schedule task at 10:00 AM on the same date at next month.

6. Schedule task at 10:00 AM tomorrow.

7. Schedule task at 10:00 AM tomorrow.

8. Schedule task to execute just after 1 hour.

9. Schedule task to execute just after 30 minutes.

10. Schedule task to execute just after 1 and 2 weeks.

11. Schedule task to execute just after 1 and 2 years.

12. Schedule task to execute at midnight.

The above job will execute on next 12:00 AM

Thanks for reading this article, I hope you will understand how to use ‘at’ command in Linux.

Related Posts

How to Scan Open Ports with Nmap

Handling filenames with spaces in Linux

How To Compare Two Files in Linux

29 Comments

EXACTLY WHAT I NEEDED.
BTW the recaptcha input is underneath the Submit button in the comments form. Hard to click.

Thanks Khay, I hope recaptcha positioned correctly now.

Hi,
does it generate log after execution like cron ?

How can we schedule a gtk “graphical” job, for example a simple yad message?
yad –title “Warning” –text “Alarm now, attention” –on-top –borders=25

In my tests, no display is shown at a specified time.

How is it better to schedule installation of upgrades?
For some reason tasks don’t execute. I tried something like: sudo apt-get upgrade -y | at 21:00

It should be
echo “apt-get upgrade -y” | sudo at 21:00

how can we Schedule one job at two different timing with single command

touch helloworld.txt | at now + 1 minute | at now + 2 hours

Where do I see the output ? How will I know if the job has been executed or not ?

please tell me that i want to run a corntab command in every last day of the month.
i am thinking but in some its 30 days and 31 in some and 28 and 29 like that.

You can try cron like this:

here [ “$(date +%d -d tomorrow)” = “01” ] will retrun true if tomorrow is the first day of next month.

I’m using at to run mplayer and stream some radio station, when i want to stop mplayer and use atrm command it doesn’t work. i got only ” Warning: deleting running job” . job vanishes from the atq list, but it still streams music.

HI, I am experiencing an issue. When i am using the format:
command | at time

It is not executing the command that I am providing.
Please tell how to get past it.

echo “command” | at time

Great article Rahul!
I just wanted to extend it a bit with some useful additions.
1) You can use -f option to point “at” to the script you need to run:
at -f /path/to/the/script time_spec
2) One can use “at” to start a process in background without nohup, etc. As easy as
at -f /a/command now
or
echo “/a/command” | at now
3) You can use “at” to run a command repeatedly, but unlike cron you can use “at” to run commands with some period between runs, for example after 3 minutes after previous run was completed. This allow you to avoid various checks preventing next run to start before previous is finished.
Moreover you can define this period as random value. Examples:
The script (lets name it /home/user1/at_run.sh):
————————————-
#!/bin/bash
/the/command/you/need
# fixed period between runs
period=3
# or random period. RANDOM is a bash’s random number from 0 to 32767
period=$[ ($RANDOM % 20) + 15 ]
at -f /home/user1/at_run.sh now + $period minutes
————————————-
run /home/user1/at_run.sh and all next runs will be scheduled automatically, so your /the/command/you/need will run repeatedly forever. Sure, you can break the next run with atq/atrm.

Thx Rahul and Sergey.

@Sergey, in your last point you are basically using a wrapper with a random pause and re-scheduling of the at job.

I think this is also achievable with cron, without the “various checks preventing next run to start before previous is finished”.
You can just replace the last line of your script with this:

sleep $(($period * 60)) && exec /home/user1/at_run.sh

Then set a crontab for /home/user1/at_run.sh .
Note: the ‘exec’ bash builtin will prevent calling bash recursively (nested bash’s) and spare memory.
You will need to kill the process to end it.

If the job is not supposed to stay permanently (boot safe), I would prefer to use an interactive bash inside a ‘screen’ command and just do a loop:

while : ; do
/the/command/you/need
sleep 60
done

So you can follow in “live” the output of the script.

hello!
i am running centos. when i submit an at nothing happens. i can call it up using “atq #”, however, it doesn’t execute?
thanks for the help.

What are the differences between at-command and chrontab?

at command is used for one-time task scheduling. crontab is used for repeated tasks.

I need to run a script at particular time-stamp which internally calls one more script.
But I am getting an error :

at `./radConnect.sh` 09:00
sh: netstat: command not found
sh: /sbin/ifconfig: No such file or directory
Radskman rc:[0] [Request successfully completed]
syntax error. Last token seen: h
Garbled time

Schedule job like following

Hi Rahul,
Thanks for the quick response, the use case which I am trying is a bit different.
I have created some containers using a script and a script internally calls another one which needs to run at a particular timestamp. So the command prompt for giving the script won’t be visible. So is there any alternative to it. Or can we do it using crontab?

Try following command.

What would be the command for future date? midnight on 30 October 2017

echo “sh radConnect.sh” | at ?

i think it should be
echo “sh radConnect.sh” | at 0 0 30 10 * *

this would repeat every year on 30th October.

I think i’ve mixed up with crontab function command.

Correction of the previous command:
echo “sh radConnect.sh” | at 12:00 AM 30.10.17

Hi Nabil.. This will work

Wow that was unusual. I just wrote an incredibly long comment but after I clicked submit my comment didn’t appear. Grrrr… well I’m not writing all that over again. Anyway, just wanted to say fantastic blog!

We are interested as well as enthusiastic about what you really are writing about here.

Источник

at Command in Linux with Examples

at command is a command-line utility that is used to schedule a command to be executed at a particular time in the future. Jobs created with at command are executed only once. The at command can be used to execute any program or mail at any time in the future. It executes commands at a particular time and accepts times of the form HH:MM to run a job at a specific time of day. The following expression like noon, midnight, teatime, tomorrow, next week, next Monday, etc. could be used with at command to schedule a job.

Syntax:

Installation of at command

For Ubuntu/Debian :

For CentOS/Fedora :

Working with at command

1. Command to list the user’s pending jobs:

2. Schedule a job for the coming Monday at a time twenty minutes later than the current time:

3. Schedule a job to run at 1:45 Aug 12 2020:

4. Schedule a job to run at 3pm four days from now:

5. Schedule a job to shutdown the system at 4:30 today:

6. Schedule a job to run five hour from now:

7. at -r or atrm command is used to deletes job , here used to deletes job 11 .

Источник

How to Use ‘at’ Command to Schedule a Task on Given or Later Time in Linux

As an alternative to cron job scheduler, the at command allows you to schedule a command to run once at a given time without editing a configuration file.

The only requirement consists of installing this utility and starting and enabling its execution:

Next, start and enable the at service at the boot time.

Once atd is running, you can schedule any command or task as follows. We want to send 4 ping probes to www.google.com when the next minute starts (i.e. if it’s 22:20:13, the command will be executed at 22:21:00) and report the result through an email ( -m , requires Postfix or equivalent) to the user invoking the command:

If you choose to not use the -m option, the command will be executed but nothing will be printed to standard output. You can, however, choose to redirect the output to a file instead.

In addition, please note that at not only allows the following fixed times: now, noon (12:00), and midnight (00:00), but also custom 2-digit (representing hours) and 4-digit times (hours and minutes).

To run updatedb at 11 pm today (or tomorrow if the current date is greater than 11 pm), do:

To shutdown the system at 23:55 today (same criteria as in the previous example applies):

You can also delay the execution by minutes, hours, days, weeks, months, or years using the + sign and the desired time specification as in the first example.

Time specifications are subject to the POSIX standard.

Summary

As a rule of thumb, use at instead of cron job scheduler whenever you want to run a command or execute a given task at a well-defined time only once. For other scenarios, use cron.

Next, we shall cover how to encrypt tar archive files using openssl, till then stay connected to Tecmint.

If You Appreciate What We Do Here On TecMint, You Should Consider:

TecMint is the fastest growing and most trusted community site for any kind of Linux Articles, Guides and Books on the web. Millions of people visit TecMint! to search or browse the thousands of published articles available FREELY to all.

If you like what you are reading, please consider buying us a coffee ( or 2 ) as a token of appreciation.

We are thankful for your never ending support.

Related Posts

5 thoughts on “How to Use ‘at’ Command to Schedule a Task on Given or Later Time in Linux”

1. 

How to add date along with time.

Hi,
Very useful article

remove extra ‘d ‘ from atdd

# chkconfig –level 35 atdd on

• 

Thanks for finding it useful, and corrected in the writeup..

Very useful, thank you!

Good article, Gabriel!

Just my 5 cents to it.

“at” can be used to start something on background, like nohup staff, but more short:
$ echo “rsync remote_resource local_resource” | at now
or
$ at -f /path/to/my/script -m now
just notice “-f script” option – it’s very convenient to run more complex staff

“at” can be used instead of cron to run something repeatedly after the first execution ends. This can eliminate cronjobs, scheduled to run each minute with various locks to avoid to run several commands simultaneously. One can add at the end of its script:
—— bash code snippet —–
# script payload here
this_script_name=$(basename $0)
this_script_path=$(readlink –canonicalize $(dirname $0))
at -f $this_script_path/$this_script_name now
—— bash code snippet —–
this will start the script again when it’s finished.

Another interesting application that’s unable to implement with cron is periodically execution with random intervals. Mostly the same as abouve, but just add random delay:
—— bash code snippet —–
# script payload here
this_script_name=$(basename $0)
this_script_path=$(readlink –canonicalize $(dirname $0))
MIN_DELAY=15
RANDOM_BIAS=20
delay=$(( ($RANDOM % $RANDOM_BIAS) + $MIN_DELAY ))
at -f $this_script_path/$this_script_name now + $delay min
—— bash code snippet —–

Got something to say? Join the discussion. Cancel reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Источник