[
https://issues.apache.org/jira/browse/FLINK-21432?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17295081#comment-17295081
]
Yang Wang commented on FLINK-21432:
-----------------------------------
Could you find the JobManager failover when the UI is not working?
If you are running standalone cluster with HA enabled, I strongly suggest to
use the pod ip, not the Kubernetes service for JobManager rpc address. Refer
here[1] for how to do this.
Address the comments in FLINK-21594.
Since the standalone cluster on K8s and native K8s shares the same Kubernetes
HA service, Flink will always add the following labels. It is harmless and just
for easier to filter the HA related config map.
configmap-type: high-availability
type: flink-native-kubernetes
If you are NOT using the {{flink run-application}} or {{kubernetes-session.sh}}
to create the Flink clusters, then it is the standalone mode.
[1].
https://stackoverflow.com/questions/66219093/flink-fencing-errors-in-k8-ha-mode/66228073#66228073
> Web UI -- Error - {"errors":["Service temporarily unavailable due to an
> ongoing leader election. Please refresh."]}
> -------------------------------------------------------------------------------------------------------------------
>
> Key: FLINK-21432
> URL: https://issues.apache.org/jira/browse/FLINK-21432
> Project: Flink
> Issue Type: Bug
> Components: Deployment / Kubernetes
> Affects Versions: 1.12.0
> Environment: debian
> Reporter: Bhagi
> Priority: Critical
> Fix For: 1.12.0
>
> Attachments: image-2021-02-22-10-39-06-180.png,
> image-2021-02-23-23-12-03-118.png, image-2021-02-23-23-14-13-708.png,
> image-2021-02-23-23-27-38-247.png
>
>
> Web UI – throwing this Error
> {"errors":["Service temporarily unavailable due to an ongoing leader
> election. Please refresh."]}
> Please find Job Manager logs.
>
> !image-2021-02-22-10-39-06-180.png!
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
