C0urante opened a new pull request, #16785: URL: https://github.com/apache/kafka/pull/16785
We've been seeing some flaky failures lately for the `MirrorConnectorsIntegrationTransactionsTest` with this stack trace: ``` org.apache.kafka.common.errors.TimeoutException: Timeout expired after 60000ms while awaiting InitProducerId ``` (We don't get any other stack frames because the Kafka `TimeoutException` class [doesn't provide them](https://github.com/apache/kafka/blob/5afdb17092f99d956bf65fcb543de0b15c04e01f/clients/src/main/java/org/apache/kafka/common/errors/ApiException.java#L45-L49)). One example of this failure can be found [here](https://ge.apache.org/s/52il7msnknzp2/tests/task/:connect:mirror:test/details/org.apache.kafka.connect.mirror.integration.MirrorConnectorsIntegrationTransactionsTest/testReplicationWithoutOffsetSyncWillNotCreateOffsetSyncsTopic()?top-execution=1). After examining the logs for one of these failures, it looks like this is either due to a broker bug, or slow CI. During a successful run of the test, it took 23 seconds for `Producer::initTransactions` to complete. It's not inconceivable that when CI load is high, it may take higher than a minute for that same method to complete, which would explain the failures we've been seeing. In order to be resilient against slow CI, or at least rule that out as a root cause, we can bump the producer's `max.block.ms` from its default of 1 minute to a higher timeout of 5 minutes, which should give it plenty of time to complete `initTransactions`. ### Committer Checklist (excluded from commit message) - [ ] Verify design and implementation - [ ] Verify test coverage and CI build status - [ ] Verify documentation (including upgrade notes) -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
