[
https://issues.apache.org/jira/browse/SOLR-18025?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18046438#comment-18046438
]
Jan Høydahl commented on SOLR-18025:
------------------------------------
So the assertion related flakiness fixed with retries already I would say is a
test issue. While the shutdown timeouts are more likely bugs in graceful
shutdown as you point out. So agree we can close this and track down the hang
situation in those other issues, though they are very old...
> LeaderTragicEventTest is flaky
> ------------------------------
>
> Key: SOLR-18025
> URL: https://issues.apache.org/jira/browse/SOLR-18025
> Project: Solr
> Issue Type: Improvement
> Components: Tests
> Reporter: Jan Høydahl
> Assignee: Jan Høydahl
> Priority: Major
> Labels: pull-request-available
> Fix For: 10.0, 9.11
>
> Time Spent: 1h
> Remaining Estimate: 0h
>
> LeaderTragicEventTest shows significant flakiness in CI builds over the last
> 28 days:
> - 42 flaky test occurrences (4% flaky rate)
> - 1,110 passed occurrences (96%)
> - Mean execution time: 8.0 seconds
>
> This test is related to LUCENE-8692 where IndexWriter.getTragicException()
> may not reflect all corrupting exceptions. The test uses
> MockDirectoryWrapper's corruptFiles to introduce corruption and expects the
> leader to give up leadership, but this doesn't always happen reliably.
> Multiple Jenkins build failures observed across Solr-Test-10.x,
> Solr-Test-10.0, and Solr-Test-main.
> Should we consider marking as @BadApple annotation?
> Test Location:
> solr/core/src/test/org/apache/solr/cloud/LeaderTragicEventTest.java
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
