[
https://issues.apache.org/jira/browse/FLINK-15691?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Chesnay Schepler closed FLINK-15691.
------------------------------------
Fix Version/s: 1.10.0
Resolution: Fixed
master: f7a7d89ca7156a57365252f947dab82455f2ff78
1.10: d7c84e6be16cfe0bf0c07baff1899f6828c02e5b
> TaskExecutorPartitionLifecycleTest#runInTaskExecutorThreadAndWait does not
> run in main thread
> ---------------------------------------------------------------------------------------------
>
> Key: FLINK-15691
> URL: https://issues.apache.org/jira/browse/FLINK-15691
> Project: Flink
> Issue Type: Bug
> Components: Runtime / Coordination, Tests
> Affects Versions: 1.9.1, 1.10.0
> Reporter: Till Rohrmann
> Assignee: Chesnay Schepler
> Priority: Critical
> Labels: pull-request-available
> Fix For: 1.10.0, 1.11.0
>
> Time Spent: 20m
> Remaining Estimate: 0h
>
> The method
> {{TaskExecutorPartitionLifecycleTest#runInTaskExecutorThreadAndWait}} does
> not run the given runnable in the {{TaskExecutors}} main thread. Instead it
> will run it in an arbitrary thread of the {{RpcService}}.
> The reason why this does not cause test instabilities is that we execute
> operations which don't cause race conditions. However, in order to avoid
> spreading false friends I would suggest to get rid of this pattern. Moreover,
> I would question whether {{TaskExecutorPartitionLifecycleTest}} really needs
> access to the {{TaskExecutors}} internal state. I believe there is an easy
> way using the {{TaskExecutors}} public API to implement the tests.
> cc [~chesnay], [~azagrebin]
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
