Uploaded image for project: 'OpenShift Bugs'
  1. OpenShift Bugs
  2. OCPBUGS-50907

Propagate Reason for Available condition from HostedControlPlane to HostedCluster

XMLWordPrintable

    • Icon: Bug Bug
    • Resolution: Unresolved
    • Icon: Undefined Undefined
    • None
    • 4.19.0
    • HyperShift
    • None
    • None
    • False
    • Hide

      None

      Show
      None
    • Hide
      *Cause*: When HostedCluster is not available, the reason is not propagated correctly from HostedControlPlane in the condition.
      *Consequence*: Only Status and Message are propagated for the "Available" condition in HostedCluster custom resource. Not the Reason.
      *Fix*: The fix propagates also the Reason, not only Message and Status.
      *Result*: The condition includes more information which helps identify the root cause of unavailability.
      Show
      *Cause*: When HostedCluster is not available, the reason is not propagated correctly from HostedControlPlane in the condition. *Consequence*: Only Status and Message are propagated for the "Available" condition in HostedCluster custom resource. Not the Reason. *Fix*: The fix propagates also the Reason, not only Message and Status. *Result*: The condition includes more information which helps identify the root cause of unavailability.
    • Bug Fix
    • In Progress

      Description of problem:

      The Reason for the Available condition should propagate from HostedControlPlane to HostedCluster together with Status and Message. Currently, only Status and Message is propagated: link. In this case we end up with KASLoadBalanderNotReachable in HCP and WaitForAvailable in HC, but we propagate the detailed message:

      ᐅ oc get hc $CLUSTER_NAME -n $NAMESPACE -oyaml

  - lastTransitionTime: "2025-02-13T15:13:55Z"
    message: 'Get "https://ad470e4971ffe4f24bb0085802628868-46f6d7fdaaca476a.elb.us-east-1.amazonaws.com:6443/healthz":
      dial tcp: lookup ad470e4971ffe4f24bb0085802628868-46f6d7fdaaca476a.elb.us-east-1.amazonaws.com
      on 172.31.0.10:53: no such host'
    observedGeneration: 3
    reason: WaitingForAvailable
    status: "False"
    type: Available

ᐅ oc get hcp hc1 -n clusters-hc1 -oyaml

  - lastTransitionTime: "2025-02-13T15:14:09Z"
    message: 'Get "https://ad470e4971ffe4f24bb0085802628868-46f6d7fdaaca476a.elb.us-east-1.amazonaws.com:6443/healthz":
      dial tcp: lookup ad470e4971ffe4f24bb0085802628868-46f6d7fdaaca476a.elb.us-east-1.amazonaws.com
      on 172.31.0.10:53: no such host'
    observedGeneration: 1
    reason: KASLoadBalancerNotReachable
    status: "False"
    type: Available

      Version-Release number of selected component (if applicable):

      4.19.0

      How reproducible:

      Reproduced as part of https://issues.redhat.com/browse/OCPBUGS-49913 which uses cluster-wide proxy for the management cluster. In this case, the HCP and HC do not become available and show the errors above.

      Steps to Reproduce:

      Steps described in this JIRA comment

      Actual results:

      The Reason for HCP is KASLoadBalancerNotReachable while the reason for HC is WaitingForAvailable, but the Message is same in both cases.

      Expected results:

      The reason KASLoadBalancerNotReachable is propagated to HostedCluster.

      Additional info:

              mgencur@redhat.com Martin Gencur
              mgencur@redhat.com Martin Gencur
              Martin Gencur Martin Gencur
              Votes:
              0 Vote for this issue
              Watchers:
              4 Start watching this issue

                Created:
                Updated: