Fedora Account System
Red Hat Associate
Red Hat Customer
Description of problem (please be detailed as possible and provide log snippests): ----------- Rbd block volume mode PVC expansion is not succeeding but waiting on 'FileSystemResizePending'. This issue is seen in OCP4.6-OCS4.5.1 cluster. PV capacity changed but capacity filed in PVC is not changing. PVC remains in condition "FileSystemResizePending" with a message "Waiting for user to (re-)start a pod to finish file system resize of volume on node." The PVC was already atached to a pod before doing the expansion. The cluster was deployed with OCS 4.5 and OCP 4.5 GA versions and then upgraded to OCS 4.5.1 and OCP 4.6 before running the test. The test case passed in OCS 4.5.1 - OCP 4.5 cluster. Test case: tests/manage/pv_services/pvc_resize/test_pvc_expansion.py::TestPvcExpand::test_pvc_expansion must-gather : http://magna002.ceph.redhat.com/ocsci-jenkins/openshift-clusters/jijoy-45-oct12/jijoy-45-oct12_20201012T123140/logs/failed_testcase_ocs_logs_1602519643/test_pvc_expansion_ocs_logs/ Test case logs : http://magna002.ceph.redhat.com/ocsci-jenkins/openshift-clusters/jijoy-45-oct12/jijoy-45-oct12_20201012T123140/logs/ocs-ci-logs-1602519643/tests/manage/pv_services/pvc_resize/test_pvc_expansion.py/TestPvcExpand/test_pvc_expansion/ PVCs which remained in 'FileSystemResizePending' when tried to expand to 25Gi NAMESPACE NAME STATUS VOLUME CAPACITY ACCESS MODES namespace-test-3a6b08f22e5840f388d8dddbc037950c pvc-test-9898596005f1418fad089a2c761863a0 Bound pvc-e1dea4aa-2ff6-47ca-8bca-543da3095a22 10Gi RWO ocs-storagecluster-ceph-rbd 6m3s namespace-test-3a6b08f22e5840f388d8dddbc037950c pvc-test-cf0afaaf6d96489e9ea6f8af722f5679 Bound pvc-9b8eba03-d479-4643-8f5d-21ea497e4529 10Gi RWX ocs-storagecluster-ceph-rbd 6m3s Status section in PVC pvc-test-9898596005f1418fad089a2c761863a0 yaml status: accessModes: - ReadWriteOnce capacity: storage: 10Gi conditions: - lastProbeTime: null lastTransitionTime: "2020-10-12T16:23:31Z" message: Waiting for user to (re-)start a pod to finish file system resize of volume on node. status: "True" type: FileSystemResizePending phase: Bound Error logs in csi-rbdplugin-bvh8m pod csi-rbdplugin container. 2020-10-12T16:24:58.777935340Z E1012 16:24:58.777895 4810 utils.go:163] ID: 67 Req-ID: 0001-0011-openshift-storage-0000000000000001-0433c7c5-0ca7-11eb-a758-0a580a81021e GRPC error: rbd: resize failed on path /var/lib/kubelet/plugins/kubernetes.io/csi/volumeDevices/publish/pvc-e1dea4aa-2ff6-47ca-8bca-543da3095a22/ce1512aa-4412-485c-8fa8-141f4ccbd12b, error: <nil> 2020-10-12T16:26:02.900019402Z E1012 16:26:02.899951 4810 utils.go:163] ID: 73 Req-ID: 0001-0011-openshift-storage-0000000000000001-0433c7c5-0ca7-11eb-a758-0a580a81021e GRPC error: rbd: resize failed on path /var/lib/kubelet/plugins/kubernetes.io/csi/volumeDevices/publish/pvc-e1dea4aa-2ff6-47ca-8bca-543da3095a22/ce1512aa-4412-485c-8fa8-141f4ccbd12b, error: <nil> Tested manually also. Seems like the condition 'FileSystemResizePending' remains forever (waited for an hour). Tried to attach the PVC which is in condition 'FileSystemResizePending' to new pod (after deleting the existing pods). The new pod did not came up running. The pod remained in ContainerCreating state. State: Waiting Reason: ContainerCreating Events: Type Reason Age From Message ---- ------ ---- ---- ------- Normal SuccessfulAttachVolume 51m attachdetach-controller AttachVolume.Attach succeeded for volume "pvc-605aec52-988f-4a31-83d2-3e2c9021ce0e" Warning FailedMapVolume 50m (x8 over 51m) kubelet, compute-0 MapVolume.MarkVolumeAsMounted failed while expanding volume for volume "pvc-605aec52-988f-4a31-83d2-3e2c9021ce0e" : Expander.NodeExpand failed to expand the volume : rpc error: code = Unknown desc = rbd: resize failed on path /var/lib/kubelet/plugins/kubernetes.io/csi/volumeDevices/publish/pvc-605aec52-988f-4a31-83d2-3e2c9021ce0e/1aa6e5b5-5d3d-489b-9837-734f5360680b, error: <nil> Warning FailedMount 11m (x17 over 49m) kubelet, compute-0 Unable to attach or mount volumes: unmounted volumes=[my-volume], unattached volumes=[default-token-8nxk9 my-volume]: timed out waiting for the condition Normal SuccessfulMountVolume 4m45s (x31 over 51m) kubelet, compute-0 MapVolume.MapPodDevice succeeded for volume "pvc-605aec52-988f-4a31-83d2-3e2c9021ce0e" volumeMapPath "/var/lib/kubelet/pods/1aa6e5b5-5d3d-489b-9837-734f5360680b/volumeDevices/kubernetes.io~csi" Normal SuccessfulMountVolume 41s (x33 over 51m) kubelet, compute-0 MapVolume.MapPodDevice succeeded for volume "pvc-605aec52-988f-4a31-83d2-3e2c9021ce0e" globalMapPath "/var/lib/kubelet/plugins/kubernetes.io/csi/volumeDevices/pvc-605aec52-988f-4a31-83d2-3e2c9021ce0e/dev" ================================================================================= Version of all relevant components (if applicable): ---------------------------------------------------- $ oc -n openshift-storage get csv NAME DISPLAY VERSION REPLACES PHASE ocs-operator.v4.5.1-594.ci OpenShift Container Storage 4.5.1-594.ci ocs-operator.v4.5.0 Succeeded $ oc get clusterversion NAME VERSION AVAILABLE PROGRESSING SINCE STATUS version 4.6.0-rc.2 True False 3h6m Cluster version is 4.6.0-rc.2 # ceph version ceph version 14.2.8-91.el8cp (75b4845da7d469665bd48d1a49badcc3677bf5cd) nautilus (stable) $ oc describe pod csi-rbdplugin-provisioner-7fc6f7bf87-2vw4l -n openshift-storage|grep Image Image: quay.io/rhceph-dev/ose-csi-external-provisioner@sha256:8a6c3a0bd6f2de58b632dbde2dd0847a8ce15ce0429a8bb9a3d5787ef38c1ea2 Image ID: quay.io/rhceph-dev/ose-csi-external-provisioner@sha256:46f51a9ccd4cf6d8543f2167fabffc34d15834ea8f023ed91eebbdbf3bfb3635 Image: quay.io/rhceph-dev/ose-csi-external-resizer@sha256:50db03c10e55f2f439ce910037ee874c9285157761b0d6b3d078d3349fa6bde3 Image ID: quay.io/rhceph-dev/ose-csi-external-resizer@sha256:1b0b15c0edf1aa3e2098b2230752b977c734b08514a1aa074fd3b5cff00b92d2 Image: quay.io/rhceph-dev/ose-csi-external-attacher@sha256:a13c104ea13944963e4ca94ae6d413a428cfd6a6d6938917a8deeb16d3c3f073 Image ID: quay.io/rhceph-dev/ose-csi-external-attacher@sha256:72e8623ee661e5120c225d7bf2c5716a74e391fee0c9cc2bd2a0e9b685d7163d Image: quay.io/rhceph-dev/cephcsi@sha256:c5ab6ca51a15066b2bdb1e9cc79afd071014147f01858f73b804750dc2376a79 Image ID: quay.io/rhceph-dev/cephcsi@sha256:795bcc3f751015e78314d7f894d8812cb49f49ff2c20d83328154fc641d0ee73 Image: quay.io/rhceph-dev/cephcsi@sha256:c5ab6ca51a15066b2bdb1e9cc79afd071014147f01858f73b804750dc2376a79 Image ID: quay.io/rhceph-dev/cephcsi@sha256:795bcc3f751015e78314d7f894d8812cb49f49ff2c20d83328154fc641d0ee73 ========================================================================== Does this issue impact your ability to continue to work with the product (please explain in detail what is the user impact)? Yes, PVC expansion is not succeeding Is there any workaround available to the best of your knowledge? No Rate from 1 - 5 the complexity of the scenario you performed that caused this bug (1 - very simple, 5 - very complex)? 1 Can this issue reproducible? Yes Can this issue reproduce from the UI? Yes If this is a regression, please provide more details to justify this: Works in OCS 4.5.0 Steps to Reproduce: 1. Create and rbd block volume mode PVC and attach it to a pod. 2. Expand the PVC and wait for the process to complete Actual results: PVC expansion pending on 'FileSystemResizePending' Expected results: PVC expansion should succeed and the Capacity field in PVC should reflect the expanded size. Additional info:
Jilju tested the image with backport PR on OCP4.6/OCS4.5, fix is working fine. Providing the dev ack. We need it to test with OCP 4.5 and OCP 4.4 as well, meanwhile I will raise the backport PR.
(In reply to Mudit Agarwal from comment #13) > Jilju tested the image with backport PR on OCP4.6/OCS4.5, fix is working > fine. Providing the dev ack. > Thanks Mudit and jilju!
PR got merged to release-4.5 branch. We will move this to ON_QA once the next OCS 4.5 build available.
(In reply to Mudit Agarwal from comment #13) > Jilju tested the image with backport PR on OCP4.6/OCS4.5, fix is working > fine. Providing the dev ack. > > We need it to test with OCP 4.5 and OCP 4.4 as well, meanwhile I will raise > the backport PR. Tested in OCP 4.5.13 - OCS 4.5.1-130.ci. Test cases passed. 1. tests/manage/pv_services/pvc_resize/test_pvc_expansion.py::TestPvcExpand::test_pvc_expansion 2. tests/manage/pv_services/pvc_resize/test_pvc_expansion.py::TestPvcExpand::test_pvc_expand_expanded_pvc
Tested in version: OCS operator v4.5.1-600.ci Cluster Version 4.6.0-0.nightly-2020-10-13-064047 Ceph Version 14.2.8-91.el8cp rook_csi_ceph cephcsi@sha256:00100822dbcdbc86286bdde8be6a459c698abd5c22d335bc121185f0a59a5b44 rook_csi_resizer ose-csi-external-resizer@sha256:50db03c10e55f2f439ce910037ee874c9285157761b0d6b3d078d3349fa6bde3 Test cases passed: tests/manage/pv_services/pvc_resize/test_pvc_expansion.py::TestPvcExpand::test_pvc_expansion tests/manage/pv_services/pvc_resize/test_pvc_expansion.py::TestPvcExpand::test_pvc_expand_expanded_pvc
(In reply to Jilju Joy from comment #20) > Tested in version: > > OCS operator v4.5.1-600.ci > Cluster Version 4.6.0-0.nightly-2020-10-13-064047 > Ceph Version 14.2.8-91.el8cp > rook_csi_ceph > cephcsi@sha256: > 00100822dbcdbc86286bdde8be6a459c698abd5c22d335bc121185f0a59a5b44 > rook_csi_resizer > ose-csi-external-resizer@sha256: > 50db03c10e55f2f439ce910037ee874c9285157761b0d6b3d078d3349fa6bde3 > > Test cases passed: > tests/manage/pv_services/pvc_resize/test_pvc_expansion.py::TestPvcExpand:: > test_pvc_expansion > tests/manage/pv_services/pvc_resize/test_pvc_expansion.py::TestPvcExpand:: > test_pvc_expand_expanded_pvc Thanks a lot Jilju for the quick verification!
Based on comment #18, comment #20 and comment #22, marking this bug as verified.