This tutorial is out of date and no longer maintained.
One of the most standard ways to run tasks in the background on Linux machines is with cron jobs. They’re useful for scheduling tasks on the VPS and automating different maintenance-related jobs. “Cron” itself is a daemon (or program) that runs in the background. The schedule for the different jobs that are run is in a configuration file called “crontab.”
Almost all distros have a form of cron installed by default. However, if you’re using a system that doesn’t have it installed, you can install it with the following commands:
For Ubuntu/Debian:
sudo apt-get update
sudo apt-get install cron
For Cent OS/Red Hat Linux:
sudo yum update
sudo yum install vixie-cron crontabs
You’ll need to make sure it runs in the background too:
sudo /sbin/chkconfig crond on
sudo /sbin/service crond start
Here is an example task we want to have run:
5 * * * * curl http://www.google.com
The syntax for the different jobs we’re going to place in the crontab might look intimidating. It’s actually a very succinct and easy-to-parse if you know how to read it. Every command is broken down into:
The command can be virtually any command you would normally run on the command line. The schedule component of the syntax is broken down into 5 different options for scheduling in the following order:
Here is a list of examples for some common schedules you might encounter while configuring cron.
To run a command every minute:
* * * * *
To run a command every 12th minute on the hour:
12 * * * *
You can also use different options for each placeholder. To run a command every 15 minutes:
0,15,30,45 * * * *
To run a command every day at 4:00am, you’d use:
0 4 * * *
To run a command every Tuesday at 4:00am, you’d use:
0 4 * * 2
You can use division in your schedule. Instead of listing out 0,15,30,45, you could also use the following:
*/4 2-6 * * *
Notice the “2-6
” range. This syntax will run the command between the hours of 2:00am and 6:00am.
The scheduling syntax is incredibly powerful and flexible. You can express just about every possible time imaginable.
Once you’ve settled on a schedule and you know the job you want to run, you’ll have to have a place to put it so your daemon will be able to read it. There are a few different places, but the most common is the user’s crontab. If you’ll recall, this is a file that holds the schedule of jobs cron will run. The files for each user are located at /var/spool/cron/crontab
, but they are not supposed to be edited directly. Instead, it’s best to use the crontab
command.
You can edit your crontab with the following command:
crontab -e
This will bring up a text editor where you can input your schedule with each job on a new line.
If you’d like to view your crontab, but not edit it, you can use the following command:
crontab -l
You can erase your crontab with the following command:
crontab -r
If you’re a privileged user, you can edit another user’s by specifying crontab -u <user> -e
For every cron job that gets executed, the user’s email address that’s associated with that user will get emailed the output unless it is directed into a log file or into /dev/null. The email address can be manually specified if you provide a “MAILTO” setting at the top of the crontab. You can also specify the shell you’d like run, the path where to search for the cron binary and the home directory with the following example:
First, let’s edit the crontab:
crontab -e
Then, we’ll edit it like so:
SHELL=/bin/bash
HOME=/
MAILTO=”example@digitalocean.com”
#This is a comment
* * * * * echo ‘Run this command every minute’
This particular job will output “Run this command every minute.” That output will get emailed every minute to the “example@digitalocean.com” email address I specified. Obviously, that might not be an ideal situation. As mentioned, we can also pipe the output into a log file or into an empty location to prevent getting an email with the output.
To append to a log file, it’s as simple as:
* * * * * echo ‘Run this command every minute’ >> file.log
Note: “>>
” appends to a file.
If you want to pipe into an empty location, use /dev/null
. Here is a PHP script that gets executed and runs in the background.
* * * * * /usr/bin/php /var/www/domain.com/backup.php > /dev/null 2>&1
Restricting access to cron is easy with the /etc/cron.allow
and /etc/cron.deny
files. In order to allow or deny a user, you just need to place their username in one of these files, depending on the access required. By default, most cron daemons will assume all users have access to cron unless one of these file exists. To deny access to all users and give access to the user tdurden, you would use the following command sequence:
echo ALL >>/etc/cron.deny
echo tdurden >>/etc/cron.allow
First, we lock out all users by appending “ALL
” to the deny file. Then, by appending the username to the allow file, we give the user access to execute cron jobs.
There are several shorthand commands you can use in your crontab file to make administering a little easier. They are essential shortcuts for the equivalent numeric schedule specified:
@hourly
- Shorthand for 0 * * * *
@daily
- Shorthand for 0 0 * * *
@weekly
- Shorthand for 0 0 * * 0
@monthly
- Shorthand for 0 0 1 * *
@yearly
- Shorthand for 0 0 1 1 *
and @reboot
, which runs the command once at startup.
Note: Not all cron daemons can parse this syntax (particularly older versions), so double-check it works before you rely on it.
To have a job that runs on start up, you would edit your crontab file (crontab -e
) and place a line in the file similar to the following:
@reboot echo "System start up"
This particular command would get executed and then emailed out to the user specified in the crontab.
Thanks for learning with the DigitalOcean Community. Check out our offerings for compute, storage, networking, and managed databases.
This textbox defaults to using Markdown to format your answer.
You can type !ref in this text area to quickly search our full set of tutorials, documentation & marketplace offerings and insert the link!
Here is a web service that makes setting up cron schedules easier: http://www.corntab.com/pages/crontab-gui
How do I put this crontab into version control? I have my git files in /src.
Hi, i have cronjob in my droplet, i set every 1hour, do some http request, process result and send to client. then after a few while, digitelocean emailed me this:
We are writing to let you know that your Droplet nest at xxx.xxx.xx.xx has been disconnected from the network after it contributed 2.1 Gbps to a 19.8 Gbps Distributed Denial of Service attack. The network traffic from your Droplet matches a pattern of malicious traffic originating from other Droplets targeted at a specific victim. We understand how disruptive this may be to your work; however, it was critical for us to disconnect your Droplet to reduce further harm.
is this process is not allowed in digitalocean? i need more explainantion. tq
Hi,
I can’t call my php script?
got this:
in crontab -e
filepath to php script is correct, and is working if i php /var/www/html/cron_test.php.
I can see the job is running, in journal ctl, but i’m getting this message:
Dec 20 19:25:01 ubuntu-blinddaters CRON[29769]: pam_unix(cron:session): session opened for user root by (uid=0) Dec 20 19:25:01 ubuntu-blinddaters CRON[29770]: (root) CMD (/usr/bin/php /var/www/html/test_cron.php) Dec 20 19:25:01 ubuntu-blinddaters CRON[29769]: (CRON) info (No MTA installed, discarding output) Dec 20 19:25:01 ubuntu-blinddaters CRON[29769]: pam_unix(cron:session): session closed for user root
Anything else in the command line i’m not sure, i’m very noob :)
help please
I try to insert UFW rules with bash using a cron job. When I run the script myself as sudo, it works fine. When the script is run as a cron job, everything works, except the insertion of the UFW rule.
In the script there are the lines:
if [ “$(whoami)” != ‘root’ ]; then echo “ufw_block.sh should be run as root. aborting” >>/var/www/hack/hack.log exit 1 fi echo “ufw_block.sh is running as root.” >>/var/www/hack/hack.log
In hack.log there appears the line “ufw_block.sh is running as root.” so I assume that the cron job runs as root.
I have tried a lot of variations of
ufw insert 1 deny from 200.200.200.200 to any
(The ip-number is just an example).
I have tried it with eval, with exec, with bash. Everything works except the insertion of the UFW rule.
Any ideas what is going wrong? Is it perhaps ufw that does not allow insertion of rules through a cron job?
According to this tutorial, */4 means it will run on 0,15,30,45.
Isn’t that false though? Doesn’t it mean it will run every 4 minutes?
Plz use my vpn
Can i run a python script using cron?
When I type “
sudo /sbin/chkconfig crond on
” it tells me that command not foundAfter a while trying to order a cron task following your tutorial and due my ignorance, I have figure out, users must write the command on a server file: nano /etc/crontab
then write the task on it: 0 4 * * * curl http://www.google.com
and after that restart the cron service: service cron restart
thanks for the detailed info