[
https://issues.apache.org/jira/browse/KAFKA-17189?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17868117#comment-17868117
]
Greg Harris commented on KAFKA-17189:
-------------------------------------
[https://docs.oracle.com/javase/7/docs/api/java/lang/Thread.html#getAllStackTraces()]
already has a liveness check, so evaluating Thread#isAlive afterwards is prone
to race conditions. If the thread is alive at the time of getAllStackTraces,
then it's leaked even if it stops executing a moment later.
> Add Thread::isAlive to DetectThreadLeak#of filter
> -------------------------------------------------
>
> Key: KAFKA-17189
> URL: https://issues.apache.org/jira/browse/KAFKA-17189
> Project: Kafka
> Issue Type: Sub-task
> Reporter: PoAn Yang
> Assignee: PoAn Yang
> Priority: Minor
>
> Although we have KAFKA-17159 to close kafka-cluster-test-kit-executor, we
> still have many thread leak when removing `waitForCondition` [0].
>
> Adding log to show whether thread is alive when showing "Thread leak
> detected" message, I found that detected thread is not alive. We may need to
> consider adding `Thread::isAlive` to `DetectThreadLeak#of` filter.
>
> [0]
> [https://ci-builds.apache.org/job/Kafka/job/kafka-pr/job/PR-16661/1/testReport/]
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
