force sync cronjob

[corneliusroemer]

• Update config, remove deployment, add revoke cronjob
• Check if having the startingDeadlineSeconds half the length of the schedule frequency helps
• Add approval step
• Also add approve to regroup and revoke cronjob
• Test out timeout
• Add timeout only to approve
• force replacement of cronjobs when sync is called
• check if removing restartpolicy will let jobs be removed by argocd
• Add back restart policy and do not throw error on timeout
• lower approve_timeout for debugging
• Just add approve_timeout to the python script
• Remove double touch
• Increase approve timeout to 5min for testing
• force a sync by adding an echo statement
• Remove reloader.stakater.com/auto: "true" just to check
• Remove echo see if we can force sync
• Empty commit
• Try out 60min

resolves #

preview URL:

Summary

Screenshot

PR Checklist

• All necessary documentation has been adapted.
• The implemented feature is covered by an appropriate test.

[anna-parker]

State at start of test

Commited small change at 10.00:
Cronjobs are scheduled to run every 30min.
State after commit is synced:

It does look like a new job is created:

However, the job cannot start as the backend is not up yet:

Oh but ingest does try again!:

And now the logs are good - we just don't see this on argocd:

Probably it will only turn green when the cronjob completes which in this case is after a new one starts.
So that means this is fine!!!!

[anna-parker]

Merged in #2473

[anna-parker]

Adding to this argocd creates a "new" pod in the same replicaset - that is also why it doesn't immediately look like the old cronjob has been stopped.