[
https://issues.apache.org/jira/browse/FLINK-14374?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16969885#comment-16969885
]
Yun Tang commented on FLINK-14374:
----------------------------------
[~gjy], adding {{FailingRestartStrategy}} is to verify FLINK-13452. If we are
sure that case would not happen with scheduler NG, we could just remove
{{FailingRestartStrategy}} directly to leave {{RegionFailoverITCase}} to verify
region failover in integrated cases as it previous did.
> Enable RegionFailoverITCase to pass with scheduler NG
> -----------------------------------------------------
>
> Key: FLINK-14374
> URL: https://issues.apache.org/jira/browse/FLINK-14374
> Project: Flink
> Issue Type: Sub-task
> Components: Runtime / Coordination, Tests
> Affects Versions: 1.10.0
> Reporter: Zhu Zhu
> Assignee: Gary Yao
> Priority: Major
> Fix For: 1.10.0
>
>
> RegionFailoverITCase currently fails with scheduler NG.
> The failure cause is that it's using {{FailingRestartStrategy}} which is not
> supported in scheduler NG.
> However, the usage of {{FailingRestartStrategy}} seems not to be necessary.
> It's for verifying a special case(see FLINK-13452) of legacy scheduler which
> are less likely to happen in the future.
> I'd propose to drop to the usage of {{FailingRestartStrategy}} in
> {{RegionFailoverITCase}} to make {{RegionFailoverITCase}} a simple test for
> streaming job on region failover.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
