Uploaded image for project: 'OCP Technical Release Team'
  1. OCP Technical Release Team
  2. TRT-2190

4.20 GCP Minor Upgrade openshift-kube-* pathological event failures

XMLWordPrintable

    • Icon: Bug Bug
    • Resolution: Done
    • Icon: Blocker Blocker
    • None
    • None
    • Incidents & Support
    • False
    • Hide

      None

      Show
      None
    • None
    • None
    • None
    • None
    • None
    • None
    • None
    • None

      Starting with 4.20.0-0.nightly-2025-07-09-121230 which is the first payload installing 4.19.0-0.nightly-2025-07-08-172809 for minor upgrades we see 4.20-upgrade-from-stable-4.19-e2e-gcp-ovn-rt-upgrade failures due to pathological events like

      event happened 25 times, something is wrong: namespace/openshift-kube-apiserver hmsg/cf99229525 - reason/SecretUpdated Updated Secret/kube-controller-manager-client-cert-key -n openshift-config-managed because it changed (05:57:30Z) result=reject 
      event happened 28 times, something is wrong: namespace/openshift-kube-apiserver hmsg/d7474cfe31 - reason/SecretUpdated Updated Secret/kube-scheduler-client-cert-key -n openshift-config-managed because it changed (05:57:05Z) result=reject 
      event happened 28 times, something is wrong: namespace/openshift-kube-apiserver hmsg/e3d56f78d6 - reason/SecretUpdated Updated Secret/control-plane-node-admin-client-cert-key -n openshift-kube-apiserver because it changed (05:54:13Z) result=reject 
      event happened 34 times, something is wrong: namespace/openshift-kube-apiserver hmsg/7727920ddf - reason/SecretUpdated Updated Secret/check-endpoints-client-cert-key -n openshift-kube-apiserver because it changed (05:57:27Z) result=reject 
      event happened 61 times, something is wrong: namespace/openshift-kube-apiserver hmsg/819d5019a3 - reason/ConfigMapUpdated Updated ConfigMap/node-system-admin-ca -n openshift-kube-apiserver-operator because it changed (05:54:17Z) result=reject 
      event happened 53 times, something is wrong: namespace/openshift-kube-apiserver hmsg/52ea701932 - reason/ConfigMapUpdated Updated ConfigMap/kube-apiserver-aggregator-client-ca -n openshift-config-managed because it changed (05:54:16Z) result=reject 
      event happened 42 times, something is wrong: namespace/openshift-kube-apiserver hmsg/1bec32f94c - reason/ConfigMapUpdated Updated ConfigMap/kube-control-plane-signer-ca -n openshift-kube-apiserver-operator because it changed (05:57:29Z) result=reject }
      

      and

      event happened 92 times, something is wrong: namespace/openshift-kube-scheduler-operator deployment/openshift-kube-scheduler-operator hmsg/6a547895ad - reason/OperatorStatusChanged Status for clusteroperator/kube-scheduler changed: Degraded message changed from "NodeControllerDegraded: All master nodes are ready" to "NodeControllerDegraded: All master nodes are ready\nResourceSyncControllerDegraded: Operation cannot be fulfilled on secrets \"kube-scheduler-client-cert-key\": the object has been modified; please apply your changes to the latest version and try again" (05:57:43Z) result=reject 
      event happened 92 times, something is wrong: namespace/openshift-kube-scheduler-operator deployment/openshift-kube-scheduler-operator hmsg/d3166e73d5 - reason/OperatorStatusChanged Status for clusteroperator/kube-scheduler changed: Degraded message changed from "NodeControllerDegraded: All master nodes are ready\nResourceSyncControllerDegraded: Operation cannot be fulfilled on secrets \"kube-scheduler-client-cert-key\": the object has been modified; please apply your changes to the latest version and try again" to "NodeControllerDegraded: All master nodes are ready" (05:57:43Z) result=reject 
      event happened 79 times, something is wrong: namespace/openshift-kube-scheduler-operator deployment/openshift-kube-scheduler-operator hmsg/64a3018a4d - reason/SecretUpdateFailed Failed to update Secret/kube-scheduler-client-cert-key -n openshift-kube-scheduler: Operation cannot be fulfilled on secrets "kube-scheduler-client-cert-key": the object has been modified; please apply your changes to the latest version and try again (05:57:43Z) result=reject }
      

      etc.

      Events shows entries like

      05:57:07 (x35)	
      
      openshift-kube-apiserver-operator	
      
      kube-apiserver-operator-cert-rotation-controller	
      
      kube-apiserver-operator	
      
      SecretUpdated
      	Updated Secret/control-plane-node-admin-client-cert-key -n openshift-kube-apiserver because it changed
      

      Which is around the time kube-apiserver clusteroperator is being upgraded

              Unassigned Unassigned
              rh-ee-fbabcock Forrest Babcock
              None
              None
              None
              None
              None
              Votes:
              0 Vote for this issue
              Watchers:
              3 Start watching this issue

                Created:
                Updated:
                Resolved: