Uploaded image for project: 'Red Hat Process Automation Manager'
  1. Red Hat Process Automation Manager
  2. RHPAM-3333

Kie Server OpenShift startup strategy watcher is closed and DC is not updated

    XMLWordPrintable

Details

    • False
    • False
    • GA
    • CR1
    • ?
    • Undefined
    • Workaround Exists
    • Hide

      When the Kie Server is not updated, restart the Kie Server pod. Then update ConfigMap to trigger a new Kie Server DC rollout.

      Show
      When the Kie Server is not updated, restart the Kie Server pod. Then update ConfigMap to trigger a new Kie Server DC rollout.
    • Hide

      Right now I do not have clear steps how to reproduce this issue. One thing that could affect and help to trigger this issue, is that the Kie Server was running for longer time (originally it was deployed to OCP 2 days ago). But as we can see sometimes this issues also as random failure in automated tests, it will be more related to updates of OpenShift ConfigMaps.

      Show
      Right now I do not have clear steps how to reproduce this issue. One thing that could affect and help to trigger this issue, is that the Kie Server was running for longer time (originally it was deployed to OCP 2 days ago). But as we can see sometimes this issues also as random failure in automated tests, it will be more related to updates of OpenShift ConfigMaps.
    • ---
    • ---
    • 2022 Week 29-31 (from Jul 18), 2022 Week 32-34 (from Aug 8), 2022 Week 35-37 (from Aug 29), 2022 Week 38-40 (from Sept 19)

    Description

      When I updated Kie Container in BC, on a Kie Server with OpenShift startup strategy. The update was done, but was not triggered DeploymentConfig redeployment. ConfigMap with Kie Server state is updated and also there is an annotation for rollout `services.server.kie.org/openshift-startup-strategy.rolloutRequired: 'true'`.
      So after this update Kie Server is not restarted with a new state.
      ConfigMap before update:

      kind: ConfigMap
      apiVersion: v1
      metadata:   annotations:     services.server.kie.org/kie-server-state.changeTimestamp: '2020-11-30T17:28:31.404572Z'
        selfLink: /api/v1/namespaces/kie-operator-pam-f126/configmaps/myapp-kieserver
        resourceVersion: '3756060'
        name: myapp-kieserver
        uid: 42475183-0ceb-438e-b6fb-94423499705d
        creationTimestamp: '2020-11-30T17:10:07Z'
        namespace: kie-operator-pam-f126
        ownerReferences:     - apiVersion: apps.openshift.io/v1
            kind: DeploymentConfig
            name: myapp-kieserver
            uid: 57d634a5-fc92-4f4e-aecf-28a12414c864
        labels:     application: myapp
          services.server.kie.org/kie-server-id: myapp-kieserver
          services.server.kie.org/kie-server-state: USED
      

      ConfigMap after update:

      kind: ConfigMap
      apiVersion: v1
      metadata:   annotations:     services.server.kie.org/kie-server-state.changeTimestamp: '2020-12-02T13:42:27.431110Z'
          services.server.kie.org/openshift-startup-strategy.rolloutRequired: 'true'
        selfLink: /api/v1/namespaces/kie-operator-pam-f126/configmaps/myapp-kieserver
        resourceVersion: '4743192'
        name: myapp-kieserver
        uid: 42475183-0ceb-438e-b6fb-94423499705d
        creationTimestamp: '2020-11-30T17:10:07Z'
        namespace: kie-operator-pam-f126
        ownerReferences:     - apiVersion: apps.openshift.io/v1
            kind: DeploymentConfig
            name: myapp-kieserver
            uid: 57d634a5-fc92-4f4e-aecf-28a12414c864
        labels:     application: myapp
          services.server.kie.org/kie-server-id: myapp-kieserver
          services.server.kie.org/kie-server-state: USED
      

      Update of Kie Server was not triggered, because on the Kie Server pod was closed OpenShift startup strategy watcher:

      �[0m�[0m19:18:35,307 INFO  [org.kie.server.services.openshift.impl.OpenShiftStartupStrategy] (OkHttp https://172.30.0.1/...) Watcher closed.
      �[0m�[0m19:18:35,308 INFO  [org.kie.server.services.openshift.impl.OpenShiftStartupStrategy] (OkHttp https://172.30.0.1/...) too old resource version: 3756060 (3818674)
      

      In attachment is full log for the Kie Server.

      We should make sure that when watcher is closed a new one is started and Kie Server is updated if needed.

      I was able to start the watcher by restarting the Kie Server pod and updating the ConfigMap again (removed and add back annotation for rollout). After this update a new deployment was created.

      Attachments

        1. log-8
          75 kB
          Jakub Schwan

        Issue Links

          Activity

            People

              rhn-support-fspolti Filippe Spolti
              jakubschwan Jakub Schwan
              Jakub Schwan Jakub Schwan
              Jakub Schwan Jakub Schwan
              Votes:
              1 Vote for this issue
              Watchers:
              7 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: