Recently, we have observed alerts like https://github.com/m-lab/dev-tracker/issues/724 due to longer datatype processing times and existing short alert wait times. We want to know quickly whether the gardener service is down. So, this change initializes the gardener_jobs_total metric at startup so that it exists without interfering with existing alerts.
Recently, we have observed alerts like https://github.com/m-lab/dev-tracker/issues/724 due to longer datatype processing times and existing short alert wait times. We want to know quickly whether the gardener service is down. So, this change initializes the
gardener_jobs_total
metric at startup so that it exists without interfering with existing alerts.This change is