Uploaded image for project: 'Kogito'
  1. Kogito
  2. KOGITO-4477

Operator sees the Infinispan as not ready after it is restarted

XMLWordPrintable

    • Icon: Bug Bug
    • Resolution: Done
    • Icon: Major Major
    • 1.4.0.Final
    • 1.2.0.Final
    • Operator
    • None
    • 2
    • False
    • False
    • Undefined
    • ---
    • ---
    • 2021 Week 07-09 (from Feb 15)

      If the Infinispan cluster is restarted, the operator is not properly determining its condition. This is due to more than one conditions available in the Infinispan YAML file and thus the check doesn't correctly obtain the status of the cluster.

      When a new Infinispan is created (either automatically using KogitoInfra or externally), its status is as follows:

      status: 
        conditions: 
          - message: 'View: kogito-infinispan-0-34441'
            status: 'True'
            type: wellFormed
      

      So there is just one condition.

      However, after it is restarted, let's say the cluster is shut down (replicas = 0) and started again (replicas = 1), its status is like this:

      status: 
        conditions: 
          - message: 'View: kogito-infinispan-0-46314'
            status: 'True'
            type: wellFormed
          - message: ''
            status: 'False'
            type: stopping
          - message: ''
            status: 'False'
            type: gracefulShutdown
      

      But in the operator code we check only the last condition in the list as there is no lastTransitionTime field. We have to check specifically for wellFormed condition as stated also in the documentation.

      This makes it difficult to deploy for example a Kogito service or Data Index service as the KogitoInfra will never show Infinispan as ready.

              mmacik@redhat.com Marian Macik
              mmacik@redhat.com Marian Macik
              Karel Suta Karel Suta
              Karel Suta Karel Suta
              Votes:
              0 Vote for this issue
              Watchers:
              1 Start watching this issue

                Created:
                Updated:
                Resolved: