-
Bug
-
Resolution: Unresolved
-
Undefined
-
None
-
CNV v4.14.10
-
None
-
0.42
-
False
-
-
False
-
CNV v4.14.11.rhel9-250
-
---
-
---
-
-
CNV I/U Operators Sprint 267
-
Moderate
-
None
Description of problem:
VMStorageClassWarning alert should fire only for Windows VMs. In theory, this was fixed in 4.14.7 with https://issues.redhat.com/browse/CNV-38482 However in CNV 4.14.9 the alert rule expression is not updated: $ oc get prometheusrule prometheus-k8s-rules-cnv -o yaml | yq -o yaml '.spec.groups[0].rules[6]' alert: VMStorageClassWarning annotations: description: '{{ $value }} Virtual Machines are in risk of causing CRC errors and major service outages' runbook_url: https://github.com/openshift/runbooks/blob/master/alerts/openshift-virtualization-operator/VMStorageClassWarning.md summary: When running VMs using ODF storage with 'rbd' mounter or 'rbd.csi.ceph.com provisioner', it will report bad crc/signature errors and cluster performance will be severely degraded if krbd:rxbounce is not set. expr: (count(kubevirt_ssp_vm_rbd_block_volume_without_rxbounce > 0) or vector(0)) > 0 labels: kubernetes_operator_component: ssp-operator kubernetes_operator_part_of: kubevirt operator_health_impact: none severity: warning
Version-Release number of selected component (if applicable):
How reproducible:
Steps to Reproduce:
1. Install cluster with CNV 4.14.9 2. Check rule: oc get prometheusrule prometheus-k8s-rules-cnv -o yaml | yq -o yaml '.spec.groups[0].rules[6]'
Actual results:
alert rule expression is checking all VMs, not only Windows VMs.
Expected results:
No regression in CNV-38482 fix
Additional info: