-
Bug
-
Resolution: Unresolved
-
Major
-
2.7.2
-
None
-
False
-
None
-
True
-
-
Description of problem:
During ""CopyDisks" phase of warm migration, the forklift controller is continuously logging following message:
level":"info","ts":"2024-10-31 11:48:49.124","logger":"plan","msg":"Did not find CDI importer pod for DataVolume.","vm":" id:vm-43263 name:'nijin-cloud-init' ","dv":"openshift-mtv/nijin-cloud-init-vm-43263-h6wxs"}
The message is wrong and the import pod was created and is downloading the disk:
# oc get pod |grep importer importer-prime-03de31bc-e6fb-4ada-bab5-063267c78f3c-checkpoint-snapshot-69124 1/1 Running 0 5m2s # oc get dv NAME PHASE PROGRESS RESTARTS AGE nijin-cloud-init-vm-43263-h6wxs ImportInProgress 24.00% 5m5s
It is looking for pod name on the original disk PVC instead of prime PVC which is why it is unable to find the import pod as per https://github.com/kubev2v/forklift/blob/e4c08befc9f3f28ae1ed2f8630e5743fad6414c7/pkg/controller/plan/migration.go#L1571
# oc get pvc prime-03de31bc-e6fb-4ada-bab5-063267c78f3c -o yaml |yq '.metadata.annotations["cdi.kubevirt.io/storage.import.importPodName"]' importer-prime-03de31bc-e6fb-4ada-bab5-063267c78f3c-checkpoint-snapshot-69124 # oc get pvc nijin-cloud-init-vm-43263-h6wxs -o yaml |yq '.metadata.annotations["cdi.kubevirt.io/storage.import.importPodName"]' null
Version-Release number of selected component (if applicable):
Migration Toolkit for Virtualization Operator 2.7.2
How reproducible:
100%
Steps to Reproduce:
Start a warm migration and while it is downloading the disk, look at the forklift-controller logs.
Actual results:
The message is invalid and can cause confusion during troubleshooting.
- links to
-
RHBA-2024:142279
MTV 2.7.4 Images
