Scheduler
Scheduler for Tiki (cron jobs) An internal scheduler was added in Tiki17, so that only one cron job needs to be defined in the system crontab, and all actions to be run a scheduled times can then be defined within Tiki UI.

1.1. How to find it
In the quick admin module, and in the main application menu under "Settings" section ("https://example.org/tiki-admin_schedulers.php").

1.2. Overview

You can list scheduled tasks

Click to expand

You can edit one scheduled task

Click to expand

Syntax for the scheduled run time is similar to the syntax used in system crontabs in GNU/Linux and equivalent operating systems:

Use CRON format to enter the values in “Run Time”:
Minute, Hour, Day of Month, Month, Day of Week
Eg. every 5 minutes: */5 * * * *

See https://en.wikipedia.org/wiki/Cron for details on Cron formats. For example:

Copy to clipboard

* * * * *
| | | | | 
| | | | | 
| | | | +---- Day of the Week   (range: 1-7, 1 = Monday)
| | | +------ Month of the Year (range: 1-12)
| | +-------- Day of the Month  (range: 1-31)
| +---------- Hour              (range: 0-23)
+------------ Minute            (range: 0-59)

To set master cron job

Manually

The master cron job must be run quite frequently to check if Tiki Scheduler must launch another operation.

Here are some examples (adjust to your server path and use case)

To run every minute

Copy to clipboard

* * * * * php /var/www/html/console.php scheduler:run

To run every 30 minutes

Copy to clipboard

*/30 * * * * php /var/www/html/console.php scheduler:run

Process Ownership

NOTE: If the tasks set up in your scheduler create cache files (which they often will) then you need to run this cron job as your webserver user, e.g. www-data, apache, nginx etc

Via Tiki Manager

Manager See instance:setup-scheduler-cron

Samples

Rebuild the Unified-Index with log

Using the Tiki Scheduler and a Console command, once you have set your master Cron Job (see above) you can automatically rebuild your Unified Index to assure your Tiki freshness. In this sample we also generate logs so we can investigate in case there were an issue.

Send notification by e-mail

in the event of job execution failure, it is possible to send an email to the admin to inform them of the failure. the reasons for the failure will be visible in the logs. if the process subsequently succeeds, another email can be sent to keep them informed.

To avoid the email being sent to all admins, you can specify the email addresses that will receive the notifications in the scheduler_users_to_notify_on_healed and scheduler_users_to_notify_on_stalled preferences.
Note: this will only work if both fields are filled in. Otherwise, all admins will receive an email.