Skip to content

Conversation

@ShivsundarR
Copy link
Contributor

What
Currently, when ShareConsumerImpl::poll() times out with an empty response, the background thread does not automatically stop fetching from the broker. The ShareConsumeRequestManager keeps sending requests until it actually receives any data.
This means we could continue fetching from the broker when we did not intend to and these records would unnecessarily start timing out sooner than we want.
This was noticed during a integ test which was previously flaky when it had an extra poll() before we started producing data.

PR fixes this by passing the pollTimeout through the ShareFetchEvent sending fetch requests as soon as the user passed timeout completes on the application.

@github-actions github-actions bot added triage PRs from the community consumer clients labels Dec 22, 2025
@AndrewJSchofield AndrewJSchofield added KIP-932 Queues for Kafka ci-approved and removed triage PRs from the community labels Dec 29, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants