[
https://issues.apache.org/jira/browse/FLINK-19358?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17450518#comment-17450518
]
Till Rohrmann commented on FLINK-19358:
---------------------------------------
Ah ok, I think I know understand the idea. Under the assumption that there is
only a single job, we don't have to match on the {{JobID}} but can simply check
whether there is a job that is recovered. In this case, the solution probably
works.
What I am wondering is how this will fit together with
[FLIP-194|https://cwiki.apache.org/confluence/x/Y4yqCw] that will introduce a
{{JobResultStore}}. With this component, it could become possible to support HA
for multi job applications. With the {{JobResultStore}} and assuming
determinism of the {{main()}} function and deterministic {{JobIDs}} I think it
should be possible to support HA.
If we now introduce a special case, then this might entail a breaking behaviour
change (zero {{JobID}} -> random {{JobID}} -> deterministic {{JobID}}).
A related question is how do we explain the limitation of a single job per
application if they want to use HA in the long term?
> when submit job on application mode with HA,the jobid will be 0000000000
> ------------------------------------------------------------------------
>
> Key: FLINK-19358
> URL: https://issues.apache.org/jira/browse/FLINK-19358
> Project: Flink
> Issue Type: Bug
> Components: Runtime / Coordination
> Affects Versions: 1.11.0
> Reporter: Jun Zhang
> Priority: Minor
> Labels: auto-deprioritized-major, usability
>
> when submit a flink job on application mode with HA ,the flink job id will be
> 00000000000000000000000000000000, when I have many jobs ,they have the same
> job id , it will be lead to a checkpoint error
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
