[
https://issues.apache.org/jira/browse/FLINK-27357?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Jesus H Christ updated FLINK-27357:
-----------------------------------
Description:
Flink on Kubernetes has High Availability services which build an external
service for the web ui access.
[https://sourcegraph.com/github.com/apache/flink@master/-/blob/flink-kubernetes/src/main/java/org/apache/flink/kubernetes/kubeclient/decorators/ExternalServiceDecorator.java]
In the case of multiple job managers to avoid 15s or so between job manager
restarts, a new job manager leader can be elected. In this case I think the
service no longer points to the correct job manager as any of the job manager
pods can have the jobmanager label.
It might help to use IPs for JobManagers like Task managers do for High
Availabiilty. It might also help to update the service with the new IP or
endpoint somehow for the external service to point to when updating the
ConfigMap for JobManager leader election:
https://sourcegraph.com/github.com/apache/flink/-/blob/flink-kubernetes/src/main/java/org/apache/flink/kubernetes/highavailability/KubernetesLeaderRetrievalDriverFactory.java
was:
Flink on Kubernetes has High Availability services which build an external
service for the web ui access.
[https://sourcegraph.com/github.com/apache/flink@master/-/blob/flink-kubernetes/src/main/java/org/apache/flink/kubernetes/kubeclient/decorators/ExternalServiceDecorator.java]
In the case of multiple job managers to avoid 15s or so between job manager
restarts, a new job manager leader can be elected. In this case I think the
service no longer points to the correct job manager as any of the job manager
pods can have the jobmanager label.
It might help to use IPs for JobManagers like Task managers do for HIgh
Availabiilty. It might also help to update the service with the new IP or
endpoint somehow for the external service to point to when updating the
ConfigMap for JobManager leader election.
> In Flink HA Service on K8S, Web UI External Service should update with new
> elected Job Manager leader's IP
> ----------------------------------------------------------------------------------------------------------
>
> Key: FLINK-27357
> URL: https://issues.apache.org/jira/browse/FLINK-27357
> Project: Flink
> Issue Type: Improvement
> Reporter: Jesus H Christ
> Priority: Minor
>
> Flink on Kubernetes has High Availability services which build an external
> service for the web ui access.
> [https://sourcegraph.com/github.com/apache/flink@master/-/blob/flink-kubernetes/src/main/java/org/apache/flink/kubernetes/kubeclient/decorators/ExternalServiceDecorator.java]
> In the case of multiple job managers to avoid 15s or so between job manager
> restarts, a new job manager leader can be elected. In this case I think the
> service no longer points to the correct job manager as any of the job manager
> pods can have the jobmanager label.
> It might help to use IPs for JobManagers like Task managers do for High
> Availabiilty. It might also help to update the service with the new IP or
> endpoint somehow for the external service to point to when updating the
> ConfigMap for JobManager leader election:
> https://sourcegraph.com/github.com/apache/flink/-/blob/flink-kubernetes/src/main/java/org/apache/flink/kubernetes/highavailability/KubernetesLeaderRetrievalDriverFactory.java
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
