Reduce number of cron retries

When the cron fails to run, that results in a large number of failed pods:

  • the CronJob's failedJobsHistoryLimit is set to 5, so 5 failed Jobs are kept;
  • the Job's backoffLimit is set to 6, so jobs that fail consistently are tried 6 times.

This means that for a single site, between 30 and 35 failed pods are around in case of consistent failure, which rather clutters the overview.

We could reduce the failedJobsHistoryLimit to 3 for example. Also we could set the backoffLimit to 1: it's not so bad when a single cron run is missed because of a temporary failure.

Edited by Arie Peterson