Every time we restart a compute resource controller, we're creating a new revision version for each job definition in AWS Batch
Maybe we should avoid doing this, and instead use always the latest version.
I think we should check whether the existing definition is consistent with the new one, and if so, do nothing. It's possible that the docker image or something else might change.
Every time we restart a compute resource controller, we're creating a new revision version for each job definition in AWS Batch Maybe we should avoid doing this, and instead use always the latest version.