Strimzi Cluster Operator failed to connect to zookeeper

Issue

Even on a fresh installed kafka cluster, using redhat shipped examples, the strimzi-cluster-operator can't talk to the zookeeper pods on :2181. This doesn't appear to be an issue until the zookeeper spec is updated which triggers the refresh and rolling restart.

[root@test cluster-operator]# oc edit Kafka my-cluster -n cg

change 

  zookeeper:
    livenessProbe:
      initialDelaySeconds: 120
      timeoutSeconds: 5
    readinessProbe:
      initialDelaySeconds: 120
      timeoutSeconds: 5
    replicas: 3
    storage:
      type: ephemeral

TO:

    livenessProbe:
      initialDelaySeconds: 60
      timeoutSeconds: 5
    readinessProbe:
      initialDelaySeconds: 60
      timeoutSeconds: 5

[root@test cluster-operator]# oc edit Kafka my-cluster -n cg
kafka.kafka.strimzi.io/my-cluster edited

... strimzi operator;

2019-05-15 03:37:20 INFO  AbstractAssemblyOperator:167 - Reconciliation #9(timer) Kafka(cg/my-cluster): Assembly my-cluster should be created or updated
2019-05-15 03:37:20 INFO  AbstractAssemblyOperator:312 - Reconciliation #9(timer) Kafka(cg/my-cluster): Assembly reconciled
2019-05-15 03:38:12 INFO  AbstractAssemblyOperator:281 - Reconciliation #10(watch) Kafka(cg/my-cluster): Kafka my-cluster in namespace cg was MODIFIED
2019-05-15 03:38:12 INFO  AbstractAssemblyOperator:167 - Reconciliation #10(watch) Kafka(cg/my-cluster): Assembly my-cluster should be created or updated
2019-05-15 03:38:12 INFO  ZookeeperLeaderFinder:90 - Trusting certificate ca.crt from Secret my-cluster-cluster-ca-cert
2019-05-15 03:38:23 WARN  ZookeeperLeaderFinder:253 - ZK my-cluster-zookeeper-0.my-cluster-zookeeper-nodes.cg.svc.cluster.local:2181: failed to connect to zookeeper:
2019-05-15 03:38:23 INFO  ZookeeperLeaderFinder:192 - No leader found for cluster my-cluster in namespace cg; backing off for 0ms (cumulative 0ms)
2019-05-15 03:38:33 WARN  ZookeeperLeaderFinder:253 - ZK my-cluster-zookeeper-0.my-cluster-zookeeper-nodes.cg.svc.cluster.local:2181: failed to connect to zookeeper:
2019-05-15 03:38:33 INFO  ZookeeperLeaderFinder:192 - No leader found for cluster my-cluster in namespace cg; backing off for 5000ms (cumulative 5000ms)

And if using the nc tool, the ZK pod can not be connected too,

[root@ndccsi-sesosm01 cluster-operator]# oc rsh strimzi-cluster-operator-6bccf4d586-n9xx8
sh-4.2$ nc -v my-cluster-zookeeper-0.my-cluster-zookeeper-client.cg.svc.cluster.local 2181
Ncat: Version 7.50 ( https://nmap.org/ncat )
Ncat: Connection timed out.
sh-4.2$ nc -v my-cluster-zookeeper-client 2181
Ncat: Version 7.50 ( https://nmap.org/ncat )
Ncat: Connection timed out.
sh-4.2$ nc -v my-cluster-zookeeper-nodes 2181
Ncat: Version 7.50 ( https://nmap.org/ncat )
Ncat: Connection to XX.XX.XX.XX failed: Connection timed out.
Ncat: Trying next address...
Ncat: Connection to XX.XX.XX.XX failed: Connection timed out.
Ncat: Trying next address...
Ncat: Connection timed out.
sh-4.2$ nc -v my-cluster-kafka-bootstrap 9092
Ncat: Version 7.50 ( https://nmap.org/ncat )
Ncat: Connected to XX.XX.XX.XX:9092.
^C
sh-4.2$ nc -v my-cluster-kafka-bootstrap 9093
Ncat: Version 7.50 ( https://nmap.org/ncat )
Ncat: Connected to XX.XX.XX.XX:9093.
^C
sh-4.2$ nc -v my-cluster-kafka-brokers 9092
Ncat: Version 7.50 ( https://nmap.org/ncat )
Ncat: Connected to XX.XX.XX.XX:9092.