[
https://issues.apache.org/jira/browse/FLINK-15434?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17009888#comment-17009888
]
Gary Yao edited comment on FLINK-15434 at 1/7/20 4:43 PM:
----------------------------------------------------------
[~hailong wang] Can you explain how you reproduced the issue locally? I ran
this test more than 10000 times without failure.
If I understand correctly, this is only a problem with the test. Your solution
would work but it could also mask out other bugs. Moreover, I think that an
AtomicReference would be more suited. Do we know why the heartbeat times out?
From what I can see, the timeout is set to only 10 ms, which seems quite low.
cc: [~trohrmann]
was (Author: gjy):
[~hailong wang] Can you explain how you reproduced the issue locally? I ran
this test more than 10000 times without failure.
If I understand correctly, this is only a problem with the test. Your solution
would work but it could also mask out other bugs. Do we know why the heartbeat
times out? From what I can see, the timeout is set to only 10 ms, which seems
quite low.
cc: [~trohrmann]
> testResourceManagerConnectionAfterRegainingLeadership test fail when run
> azure
> -------------------------------------------------------------------------------
>
> Key: FLINK-15434
> URL: https://issues.apache.org/jira/browse/FLINK-15434
> Project: Flink
> Issue Type: Bug
> Components: Tests
> Affects Versions: 1.9.1
> Reporter: hailong wang
> Priority: Critical
> Fix For: 1.10.0
>
>
> Error message
> Expected: <b65797325db6323dd5f8bdfeeaa18467>
> but: was <57f975298d116a9d7623f5a844ce6502>
>
> Stack trace
> java.lang.AssertionError: Expected: <b65797325db6323dd5f8bdfeeaa18467> but:
> was <57f975298d116a9d7623f5a844ce6502> at
> org.apache.flink.runtime.jobmaster.JobMasterTest.testResourceManagerConnectionAfterRegainingLeadership(JobMasterTest.java:1033)
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
