[
https://issues.apache.org/jira/browse/KAFKA-20112?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18064473#comment-18064473
]
Lianet Magrans commented on KAFKA-20112:
----------------------------------------
Haven't had the time to look into this in detail, but seems the failure comes
from a timeout when closing the consumer implicitly with the try (not from the
core logic being tested, which is the revocation callback on a rebalance). It
will probably be helpful to try to repro with logging enabled for
OffsetsRequestManager#updateFetchPositions, to understand why the consumer may
not be getting position in time on close in some cases (maybe position became
invalid?). The close stops many things (e.g., stops find coord at some point),
so I wonder if it may be a funny interaction, that we end up needing something
on close that we cannot resolve anymore. Logs on the func above should let us
see where we get stuck (if we can repro locally)
> Flaky testAsyncConsumerMaxPollIntervalMsDelayInRevocation
> ---------------------------------------------------------
>
> Key: KAFKA-20112
> URL: https://issues.apache.org/jira/browse/KAFKA-20112
> Project: Kafka
> Issue Type: Test
> Components: clients, consumer
> Reporter: Lianet Magrans
> Priority: Major
>
> Has been flaky on trunk
> https://develocity.apache.org/scans/tests?search.relativeStartTime=P28D&search.rootProjectNames=kafka&search.tags=trunk&search.timeZoneId=America%2FToronto&tests.container=org.apache.kafka.clients.consumer.PlaintextConsumerPollTest&tests.sortField=FLAKY&tests.test=testAsyncConsumerMaxPollIntervalMsDelayInRevocation()%5B1%5D
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
