-
Bug
-
Resolution: Duplicate
-
Normal
-
None
-
Quality / Stability / Reliability
-
False
-
False
-
CLOSED
-
Important
-
No
Description of problem: After 4.9.3->4.10.0 upgrade, intermittently 1 vmim is reporting to be in failed state.
Version-Release number of selected component (if applicable):
4.9.3->4.10.0-662 cnv upgrade
How reproducible:
intermittent
Steps to Reproduce:
1. Run upgrade tests and watch vmims post upgrade
2.
3.
Actual results:
the vm has associated poddisruptionbudget and it was migration eligible in terms of available resource on target
vmim:
=========
[cnv-qe-jenkins@c01-cnv410-upg-2dx9x-executor ~]$ kubectl get vmim -A | grep nfs
upgrade-operators-product-upgrade-test-upgrade kubevirt-evacuation-lvx9l Succeeded vm-for-product-upgrade-nfs-1644416165-7499158
upgrade-operators-product-upgrade-test-upgrade kubevirt-evacuation-nbzvj Succeeded vm-for-product-upgrade-nfs-1644416165-7499158
upgrade-operators-product-upgrade-test-upgrade kubevirt-workload-update-sm74p Failed vm-for-product-upgrade-nfs-1644416165-7499158
[cnv-qe-jenkins@c01-cnv410-upg-2dx9x-executor ~]$
===================
from the controller logs:
================
.....
{"component":"virt-controller","kind":"","level":"info","msg":"no migration policy matched for VMI vm-for-product-upgrade-nfs-1644416165-7499158","name":"vm-for-product-upgrade-nfs-1644416165-7499158","namespace":"upgrade-operators-product-upgrade-test-upgrade","pos":"migration.go:1576","timestamp":"2022-02-09T15:33:17.125098Z","uid":"2fb6613b-ed53-46b4-851b-9b8430d415eb"} {"component":"virt-controller","kind":"","level":"info","msg":"Handed off migration upgrade-operators-product-upgrade-test-upgrade/kubevirt-workload-update-sm74p to target virt-handler.","name":"vm-for-product-upgrade-nfs-1644416165-7499158","namespace":"upgrade-operators-product-upgrade-test-upgrade","pos":"migration.go:683","timestamp":"2022-02-09T15:33:17.168461Z","uid":"2fb6613b-ed53-46b4-851b-9b8430d415eb"} {"component":"virt-controller","kind":"","level":"error","msg":"vmi's migration state was taken over by another migration object","name":"kubevirt-workload-update-sm74p","namespace":"upgrade-operators-product-upgrade-test-upgrade","pos":"migration.go:427","timestamp":"2022-02-09T15:33:17.185284Z","uid":"433e4369-fd4b-4947-9a0f-34fd90871367"}.....
{"component":"virt-controller","kind":"","level":"info","msg":"shrinking pdb upgrade-operators-product-upgrade-test-upgrade/kubevirt-disruption-budget-5vb6v due to migration completion","name":"vm-for-product-upgrade-nfs-1644416165-7499158","namespace":"upgrade-operators-product-upgrade-test-upgrade","pos":"disruptionbudget.go:528","timestamp":"2022-02-09T15:33:39.616025Z","uid":"2fb6613b-ed53-46b4-851b-9b8430d415eb"}==================
vmi however points to updated image:
==================
launcherContainerImageVersion: registry.redhat.io/container-native-virtualization/virt-launcher@sha256:7e52dd9a08df07e909a729a941e6360250ab95fc3ee06c04c3d74537aed9f513
==================
cross referenced it with csv.spec.relatedImages and other migrated vms.
[cnv-qe-jenkins@c01-cnv410-upg-2dx9x-executor ~]$ kubectl get csv kubevirt-hyperconverged-operator.v4.10.0 -n openshift-cnv -o json | jq ".spec.relatedImages" | grep launcher
"image": "registry.redhat.io/container-native-virtualization/virt-launcher@sha256:7e52dd9a08df07e909a729a941e6360250ab95fc3ee06c04c3d74537aed9f513",
"name": "registry.redhat.io/container-native-virtualization/virt-launcher:v4.10.0-210"
[cnv-qe-jenkins@c01-cnv410-upg-2dx9x-executor ~]$
Details of the issue: https://bugzilla.redhat.com/show_bug.cgi?id=2021992#c15
Expected results:
Additional info:
- duplicates
-
CNV-15042 [2026357] Migration in sequence can be reported as failed even when it succeeded
-
- Closed
-
- external trackers