-
Bug
-
Resolution: Unresolved
-
Major
-
None
-
2.9.4
-
Quality / Stability / Reliability
-
False
-
-
True
-
-
-
Important
Description of problem:
After archiving and deleting the migration plan, the original and prime PVC remains in the cluster and needs to be manually removed. Only the populator pods are cleaned up properly.
The following PVCs remained in the cluster after archiving and deleting a failed plan for a VM with two disks:
42544ee2-e9af-4811-916b-3ab44a0ddf0c-nxwgq Pending ocs-external-storagecluster-ceph-rbd <unset> 19m db2f38c1-20db-48eb-8d0d-f74bc4258f73-hkvzn Pending ocs-external-storagecluster-ceph-rbd <unset> 19m prime-03c14633-d9b6-459b-a262-af216746b6c8 Bound pvc-4ff08e9e-b360-42f5-baf5-0b86bddf0abc 100Gi RWX ocs-external-storagecluster-ceph-rbd <unset> 19m prime-eae92085-2b5d-47f0-8286-c77e793ed8d5 Bound pvc-0911a693-f8ea-4b27-939d-657847d008eb 200Gi RWX ocs-external-storagecluster-ceph-rbd <unset> 19m
To delete the PVCs, we also need to remove the finalizer after deleting them.
# oc delete pvc 42544ee2-e9af-4811-916b-3ab44a0ddf0c-nxwgq db2f38c1-20db-48eb-8d0d-f74bc4258f73-hkvzn persistentvolumeclaim "42544ee2-e9af-4811-916b-3ab44a0ddf0c-nxwgq" deleted persistentvolumeclaim "db2f38c1-20db-48eb-8d0d-f74bc4258f73-hkvzn" deleted ^C[ stuck in terminating: 42544ee2-e9af-4811-916b-3ab44a0ddf0c-nxwgq Terminating ocs-external-storagecluster-ceph-rbd <unset> 21m db2f38c1-20db-48eb-8d0d-f74bc4258f73-hkvzn Terminating ocs-external-storagecluster-ceph-rbd <unset> 21m # oc patch pvc db2f38c1-20db-48eb-8d0d-f74bc4258f73-hkvzn -p '{"metadata":{"finalizers":null}}' persistentvolumeclaim/db2f38c1-20db-48eb-8d0d-f74bc4258f73-hkvzn patched # oc patch pvc 42544ee2-e9af-4811-916b-3ab44a0ddf0c-nxwgq -p '{"metadata":{"finalizers":null}}' persistentvolumeclaim/42544ee2-e9af-4811-916b-3ab44a0ddf0c-nxwgq patched # oc delete pvc prime-03c14633-d9b6-459b-a262-af216746b6c8 prime-eae92085-2b5d-47f0-8286-c77e793ed8d5 persistentvolumeclaim "prime-03c14633-d9b6-459b-a262-af216746b6c8" deleted persistentvolumeclaim "prime-eae92085-2b5d-47f0-8286-c77e793ed8d5" deleted
Version-Release number of selected component (if applicable):
Migration Toolkit for Virtualization Operator 2.9.4 OpenShift Virtualization 4.19.6
How reproducible:
Always
Steps to Reproduce:
1. Start a migration plan from RHV provider. 2. While populator pod is downloading the disk, stop the ovirt-imageio service on the RHV hosts. 3. Wait for it to fail. 4. Archive and delete the migration plan. 5. The created PVCs for the migration are not automatically removed.
Actual results:
Archiving and deleting a failed RHV plan is not cleaning up the PVC
Expected results:
It should cleanup the PVC since the leftover can cause operational issues
Additional info:
