[
https://issues.apache.org/jira/browse/FLINK-21455?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17301593#comment-17301593
]
Chesnay Schepler commented on FLINK-21455:
------------------------------------------
The JobManager (which contains the Dispatcher) already maintains a history of
jobs that have been run.
The HistoryServer also does that, but it's primary use-cases are to a) be
independent from Flink clusters b) aggregate jobs across clusters.
Because of that, a HistoryServer being linked to the lifetime of a (single)
Flink cluster just doesn't really make sense.
> Deploy history server on kubernetes
> ------------------------------------
>
> Key: FLINK-21455
> URL: https://issues.apache.org/jira/browse/FLINK-21455
> Project: Flink
> Issue Type: Improvement
> Components: Deployment / Kubernetes
> Reporter: Yi Tang
> Priority: Minor
> Labels: starter
>
> Can we deploy history server on kubernetes while starting a cluster? Maybe
> add an option?
> Looks like it can only be deployed manually for now.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
