Uploaded image for project: 'OpenShift Bugs'
  1. OpenShift Bugs
  2. OCPBUGS-7540

TALM may miss MCP reconcile after change to PerformanceProfile or operator upgrade

XMLWordPrintable

    • No
    • CNF RAN Sprint 232, CNF RAN Sprint 233, CNF RAN Sprint 234
    • 3
    • Rejected
    • False
    • Hide

      None

      Show
      None

      Description of problem:

      When upgrading the PAO operator from 4.9 to 4.10 version the CGU showed “complete” ~10 minutes before the master MCP finished reconciling due to change in performance profile MachineConfig.

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

      4.11.3

      How reproducible:

      Always

      Steps to Reproduce:

      This procedure emulates the upgrade of a SNO OCP followed by operators.
      1. Install cluster with OCP 4.10 and operators taken from the 4.9 catalog.
      2. Create a new validator policy which contains (at least) the MCP content from the group-du-validator policy used during installation. Ensure the bindingRules apply this to your cluster.
      3. Update common-config: set catalogsource index to 4.10.
      4. Update common-subscriptions: Set PAO channel to 4.10
      5. Commit and push changes to git
      6. Create a CGU to remediate common-config, common-subscriptions, and update-validator (in that order).

      Actual results:

      The CGU shows "completed" ~10 minutes before the master MCP finishes updating with new rendered master.

      Expected results:

      CGU shows "completed" after the MCP completes its reconciliation. The update-validator policy should stay non-compliant until that reconciliation completes, however it is seen as compliant 
      

      before

       the MCP starts updaing.

      Additional info:

       

       

            saskari@redhat.com Saeid Askari
            rhn-support-imiller Ian Miller
            Bonnie Block Bonnie Block
            Nabeel Cocker
            Votes:
            0 Vote for this issue
            Watchers:
            13 Start watching this issue

              Created:
              Updated:
              Resolved: