Closed IvoryC closed 4 years ago
this was fixed on branch 'dev' on 6 July 2020. " If no job id is returned, handle that as a failure not a success." The fix is untested, but it is simple; and not worth a lot of time. This introduced the cluster.returnsBatchIds property.
When pipeline.env=cluster there is a MAIN script to submit jobs. When jobs are run in docker or on the cluster, then the job returns an id fo subsequent tracking. MAIN should not say _Success if it did not get an id back for one or more of the submitted jobs.