HA Management is STOPPED error after upgrading Db2 Warehouse on OpenShift

Condition

Wolverine is a high-availability framework that monitors and recovers services in case of component failures within a container, such as Db2® Warehouse. After an upgrade, if you run the Wolverine command wvcli system status within the Db2 Warehouse pod, you might see a message stating that the High Availability (HA) Management is STOPPED:
HA Management is STOPPED

Remedy

To resolve this issue, you need to bring Wolverine back to a health and a started state.

Procedure

  1. Exec into the Db2 pod.
  2. Stop the Wolverine high availability monitoring process: 
    wvcli system stop
  3. Remove the /newsystem etcd keystore:
    Fore Db2uCluster deployments:
    1. Find the name of the Db2ucluster custom resource:
      oc get db2ucluster --all-namespaces
    2. Using the returned value, create an environment variable for your db2ucluster resource:
      db2ucluster=<db2ucluster-resource-name>
      where <db2cluster-resource-name> is the returned value.
    3. Find the name of the etcd pod for your database deployment:
      etcd_podname=$(oc get pod --show-labels=true --selector app=${db2ucluster} | grep -i etcd-0 | awk {'print $1'})
    4. Remove the /newsystem etcd keystore:
      oc rsh ${etcd_podname} etcdctl rm -r /newsystem
    For Db2uInstance deployments:
    1. Find the name of the Db2uInstance custom resource:
      oc get db2uinstance --all-namespaces
    2. Using the returned value, create an environment variable for your Db2uInstance resource:
      db2uinstance=<db2uinstance-resource-name>
      where <db2instance-resource-name> is the returned value.
    3. Find the name of the etcd pod for your database deployment:
      etcd_podname=$(oc get pod --show-labels=true --selector app=${db2uinstance} | grep -i etcd-0 | awk {'print $1'})
    4. Remove the /newsystem etcd keystore:
      oc rsh ${etcd_podname} etcdctl rm -r /newsystem
  4. Restart the etcd pod: 
    oc delete pod ${etcd_podname}
  5. Restart the Wolverine high availability monitoring process by following step 1 of Exec into the Db2 pod and then running the following command to start the system: 
    wvcli system start