If the Pull Request Commit Backfill Job completely fails it is required to have Bitbucket restarted to trigger it again.

The problem arises if you have a few bad repositories and need to perform a manual workaround to remove them, where you need to wait for the job to fail and then capture the repository under <bitbucket-home>/shared/config/upgrades/core-backfill-pull-request-commits to have it skipped, where you need to restart Bitbucket every single time.

Please add an easier way of restarting this job.