Description of problem: In case vddk-init-image is not added to the v2v-vmware configMap, or set bug not accessible, the DataVolume doesn't report "AwaitingVDDK", like it used to on previous CNV versions. Version-Release number of selected component (if applicable): CNV-2.6.3-25 MTV-2.0.0 upstream latest May 24 2021. Additional info: This is related to Bug 1902028 - [MTV][RFE] Add a "validation" stage to the current VM migration pipeline. This last bug propagates the CDI importer DataVolume reported reason to the migration plan custom resource. and now that the "AwaitingVDDK" is missing no reason is propagated and it is not clear to why the migration plan in not progressing.
Moving the bug to modified since the PR is merged.
Matthew, On OCP-4.7/CNV-2.6.6 (iib-88835 hco-v2.6.6-35), if vddk-init-image is not configured in the v2v-vmware configMap, When running VM import from VMware using VMIO, In UI show progress 10%. There is no cdi importer pos created. cdi-deployment log shows: {"level":"error","ts":1626157204.5167053,"logger":"controller","msg":"Reconciler error","controller":"import-controller","name":"d3ab754d-c447-42bb-afde-73d0856f3626-2000","namespace":"default","error":"Found v2v-vmware ConfigMap in namespace openshift-cnv, but it does not contain a 'vddk-init-image' entry.","errorVerbose":"Found v2v-vmware ConfigMap in namespace openshift-cnv, but it does not contain a 'vddk-init-image' entry.\nkubevirt.io/containerized-data-importer/pkg/controller.(*ImportReconciler).getVddkImageName\n\t dv status shows: status: conditions: - lastHeartbeatTime: "2021-07-13T06:14:30Z" lastTransitionTime: "2021-07-13T06:14:30Z" message: PVC d3ab754d-c447-42bb-afde-73d0856f3626-2000 Bound reason: Bound status: "True" type: Bound - lastHeartbeatTime: "2021-07-13T06:14:30Z" lastTransitionTime: "2021-07-13T06:14:30Z" status: "False" type: Ready - lastHeartbeatTime: "2021-07-13T06:14:30Z" lastTransitionTime: "2021-07-13T06:14:30Z" status: "False" type: Running phase: ImportScheduled progress: N/A Is this the expected behavior please? Shouldn't we have the importer pod created?
I think this was just because the cluster was running v2.6.6-35 and the fix is in v2.6.6-38. I couldn't find any trace of the AwaitingVDDK annotation in its CDI controller binary. For clarification, the importer pod will not be created because this failure happens during pod creation.
Verified on OCP-4.7.20/CNV-2.6.6-39. If vddk-init-image is missing in the v2v-vmware config map the dv (attached) contains the expected reason: AwatingVDDK The VM import CR (attached) however shows "Copying disks" - Need to check this.
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: OpenShift Virtualization 2.6.6 Images security and bug fix 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-2021:3119