[ 
https://issues.apache.org/jira/browse/CAMEL-26638?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Claus Ibsen resolved CAMEL-26638.
---------------------------------
    Resolution: Fixed

Merged to main in https://github.com/apache/camel/pull/29248

_Claude Code on behalf of davsclaus_

> camel-hazelcast - 
> HazelcastReliableTopicConsumerLossTest.listenerCancelledByHazelcastIsReported 
> is flaky
> --------------------------------------------------------------------------------------------------------
>
>                 Key: CAMEL-26638
>                 URL: https://issues.apache.org/jira/browse/CAMEL-26638
>             Project: Camel
>          Issue Type: Test
>          Components: camel-hazelcast
>            Reporter: Claus Ibsen
>            Assignee: Claus Ibsen
>            Priority: Major
>             Fix For: 4.24.0
>
>
> HazelcastReliableTopicConsumerLossTest.listenerCancelledByHazelcastIsReported 
> fails now and then on CI (all 3 surefire attempts, for example in the PR 
> #28874 build), while it passes locally.
> The test destroys the reliable topic right after the route starts. Hazelcast 
> only cancels a listener whose ringbuffer read is parked when the topic is 
> destroyed: ProxyServiceImpl cancels the parked operations with a 
> DistributedObjectDestroyedException, and the MessageRunner then cancels the 
> listener. When the first read of the listener has not reached the member yet 
> (a loaded CI machine), it runs after the destroy, finds no ringbuffer and 
> waits on a new empty one, so the listener is never cancelled and the test 
> waits for nothing.
> Fix: destroy the topic again until the consumer reports it, so a read that 
> arrived after a destroy is cancelled by the next one. The consumer code is 
> not changed.
> _Claude Code on behalf of Claus Ibsen (davsclaus)_



--
This message was sent by Atlassian Jira
(v10.3.26#10030026)

Reply via email to