Cron jobs allow you to run scheduled tasks at specified times or intervals. The
crons section of
.platform.app.yaml describes these tasks and the schedule when they are triggered. Each item in the list is a unique name identifying a separate cron job. Crons are started right after build phase.
It has a few subkeys listed below:
- spec: The cron specification. For example:
*/19 * * * *to run every 19 minutes.
- cmd: The command that is executed, for example
cd public ; drush core-cron
The minimum interval between cron runs is 5 minutes, even if specified as less. Additionally, a variable delay is added to each cron job in each project in order to prevent host overloading should every project try to run their nightly tasks at the same time. Your crons will not run exactly at the time that you specify, but will be delayed by 0-300 seconds.
A single application container may have any number of cron tasks configured, but only one may be running at a time. That is, if a cron task fires while another cron task is still running it will pause and then continue normally once the first has completed.
Cron runs are executed using the dash shell, not the bash shell used by normal SSH logins. In most cases that makes no difference but may impact some more involved cron scripts.
If an application defines both a
web instance and a
worker instance, cron tasks will be run only on the
Cron log output is captured in the at
/var/log/cron.log. See the Log page for more information on logging.
The following example runs Drupal’s normal cron hook every 19 minutes, using Drush. It also sets up a second cron task to run Drupal’s queue runner on the aggregator_feeds queue every 7 minutes.
crons: # Run Drupal's cron tasks every 19 minutes. drupal: spec: '*/19 * * * *' cmd: 'cd web ; drush core-cron' # But also run pending queue tasks every 7 minutes. # Use an odd number to avoid running at the same time as the `drupal` cron. drush-queue: spec: '*/7 * * * *' cmd: 'cd web ; drush queue-run aggregator_feeds'