[
https://issues.apache.org/jira/browse/FLINK-32421?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Matthias Pohl resolved FLINK-32421.
-----------------------------------
Fix Version/s: 1.18.0
1.16.3
1.17.2
Assignee: Matthias Pohl
Resolution: Fixed
master: 5dbbc695ed90241bc22d01d05f19e2bdfb4b1e0b
1.17: 067df7b9ce14067375adffcc05bab5513592d646
1.16: 56cb5442333413e7617b12e5796c806d64de62e9
> EmbeddedLeaderServiceTest.testConcurrentRevokeLeadershipAndShutdown is not
> properly implemented
> -----------------------------------------------------------------------------------------------
>
> Key: FLINK-32421
> URL: https://issues.apache.org/jira/browse/FLINK-32421
> Project: Flink
> Issue Type: Bug
> Components: Runtime / Coordination, Tests
> Affects Versions: 1.16.2, 1.18.0, 1.17.1
> Reporter: Matthias Pohl
> Assignee: Matthias Pohl
> Priority: Major
> Labels: pull-request-available
> Fix For: 1.18.0, 1.16.3, 1.17.2
>
>
> The purpose of
> {{EmbeddedLeaderServiceTest.testConcurrentRevokeLeadershipAndShutdown}} is to
> check that there is no {{NullPointerException}} happening if the event
> processing happens after the shutdown of the {{EmbeddedExecutorService}} (see
> FLINK-11855).
> But the concurrent execution is not handled properly. The test also succeeds
> if the close call happened before the shutdown (due to the multi-threaded
> nature of the test) which leaves us without the actual test scenario being
> tested.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
