zookeeper-operator: Resource Status not updated

When deploying a zookeeper cluster with 0.2.0 zookeeper-operator, the ZookeeperCluster resource does not get the Status section updated and always stays at

Status:
  External Client Endpoint:  
  Internal Client Endpoint:  
  Members:
    Ready:         <nil>
    Unready:       <nil>
  Ready Replicas:  0
  Replicas:        0

Example: We’ve deployed a zk-cluster in namespace test-project that is supposed to have three replicas. Inspecting the related stateful set reveals that there are indeed three replicas running in the system:

$kubectl get statefulsets -n test-project
NAME        DESIRED   CURRENT   AGE
zookeeper   3         3         53m

Inspecting the ZookeeperCluster existing in the project, we get the Status section showing zero replicas.

$kubectl describe zk zookeeper -n test-project
Name:         zookeeper
Namespace:    test-project
Labels:       <none>
Annotations:  <none>
API Version:  zookeeper.pravega.io/v1beta1
Kind:         ZookeeperCluster
Metadata:
  Creation Timestamp:  2019-03-01T19:08:21Z
  Generation:          1
  Owner References:
    API Version:           nautilus.dellemc.com/v1alpha1
    Block Owner Deletion:  true
    Controller:            true
    Kind:                  Project
    Name:                  test-project
    UID:                   60c9fb21-3c55-11e9-a64e-005056bdd0fa
  Resource Version:        620540
  Self Link:               /apis/zookeeper.pravega.io/v1beta1/namespaces/test-project/zookeeper-clusters/zookeeper
  UID:                     6117246e-3c55-11e9-a64e-005056bdd0fa
Spec:
  Config:
    Init Limit:  10
    Sync Limit:  2
    Tick Time:   2000
  Image:
    Pull Policy:  Always
    Repository:   emccorp/zookeeper
    Tag:          3.5.4-beta-operator
  Labels:
    App:      zookeeper
    Release:  zookeeper
  Persistence:
    Access Modes:
      ReadWriteOnce
    Data Source:  <nil>
    Resources:
      Requests:
        Storage:  20Gi
  Pod:
    Affinity:
      Pod Anti Affinity:
        Preferred During Scheduling Ignored During Execution:
          Pod Affinity Term:
            Label Selector:
              Match Expressions:
                Key:       app
                Operator:  In
                Values:
                  zookeeper
            Topology Key:  kubernetes.io/hostname
          Weight:          20
    Labels:
      App:      zookeeper
      Release:  zookeeper
    Resources:
    Termination Grace Period Seconds:  30
  Ports:
    Container Port:  2181
    Name:            client
    Container Port:  2888
    Name:            quorum
    Container Port:  3888
    Name:            leader-election
  Replicas:          3
  Size:              3
Status:
  External Client Endpoint:  
  Internal Client Endpoint:  
  Members:
    Ready:         <nil>
    Unready:       <nil>
  Ready Replicas:  0
  Replicas:        0
Events:            <none>

About this issue

  • Original URL
  • State: closed
  • Created 5 years ago
  • Comments: 15 (15 by maintainers)

Most upvoted comments

@adrianmo and @Tristan1900 Thank you for the help!

I’m closing the issue. The solution is to update the CRD spec as in https://github.com/pravega/zookeeper-operator/blob/master/deploy/crds/zookeeper_v1beta1_zookeepercluster_crd.yaml#L36 and use a cluster with 1.11.x Kubernetes version.