Re-running "release allocation" job should be fail-proof. #35
Description
We don’t need to check the status of the release allocation job: if it fails because the allocation has already been released, then it should not exit with failure status.
{{ message }}