Description of problem (please be detailed as possible and provide log snippests): During uninstall process rook closes the monitoring channel for daemons. It may attempt close on closed channel leading to a panic and restart of the rook operator pod. Version of all relevant components (if applicable): 4.6
Upstream PR https://github.com/rook/rook/pull/6369
@Talur, We haven't seen this issue in our test beds. can you please share the steps to reproduce?
Added PVC from UI, deleted storagecluster from UI - storagecluster stuck on deleting, panic in the rook-ceph-operator pod is not seen [asandler@redhat ~]$ oc get storagecluster NAME AGE PHASE EXTERNAL CREATED AT VERSION ocs-storagecluster 8h Deleting 2020-10-07T17:51:22Z 4.6.0 [asandler@redhat ~]$ oc logs -n openshift-storage rook-ceph-operator-57459f5464-8fcfk | grep panic [asandler@redhat ~]$ oc get csv NAME DISPLAY VERSION REPLACES PHASE ocs-operator.v4.6.0-116.ci OpenShift Container Storage 4.6.0-116.ci Succeeded is it ok that the storagecluster is not deleted and stuck? (p.s. didn't do all uninstall procedure from start - only deleteing storage cluster from UI)
Created attachment 1719872 [details] rook-ceph-operator log
(In reply to Anna Sandler from comment #10) > Added PVC from UI, deleted storagecluster from UI - storagecluster stuck on > deleting, panic in the rook-ceph-operator pod is not seen > [asandler@redhat ~]$ oc get storagecluster > NAME AGE PHASE EXTERNAL CREATED AT > VERSION > ocs-storagecluster 8h Deleting 2020-10-07T17:51:22Z 4.6.0 > [asandler@redhat ~]$ oc logs -n openshift-storage > rook-ceph-operator-57459f5464-8fcfk | grep panic > [asandler@redhat ~]$ oc get csv > NAME DISPLAY VERSION > REPLACES PHASE > ocs-operator.v4.6.0-116.ci OpenShift Container Storage 4.6.0-116.ci > Succeeded > > is it ok that the storagecluster is not deleted and stuck? (p.s. didn't do > all uninstall procedure from start - only deleteing storage cluster from UI) Yes, the delete request will be stuck because of the graceful delete feature. I confirmed by looking at the attached logs. Relevant line - ""ocs-storagecluster-cephblockpool" has rbd images: pool "ocs-storagecluster-cephblockpool" contains images/snapshosts" Once you delete the rbd/fs PVCs, the delete request should proceed. The panic should have occurred already if the bug was not fixed.
the bug is not seen anymore. moving to verified
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (Moderate: Red Hat OpenShift Container Storage 4.6.0 security, bug fix, enhancement update), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHSA-2020:5605