Steps to reproduce: #1 Run job to backup namespace that doesn't exist can be done directly from velero or from the GUI and using the form or yaml views ex: oc exec -n openshift-adp $(oc get pods -n openshift-adp -o name | grep pod/velero) – ./velero backup create backup-cr --include-namespaces doesnt-exist #2 After job "partially fails" attempt to remove the job using a deletebackuprequest job or by rsh'ing to the velero pod and issuing the delete directly ex: ./velero backup delete backup-cr Additional Note: The backup gets deleted when data from backupstoragelocation is directly deleted and then further performing deletion of backup CR.