lhotari opened a new issue, #23505: URL: https://github.com/apache/pulsar/issues/23505
### Search before asking - [X] I searched in the [issues](https://github.com/apache/pulsar/issues) and found nothing similar. ### Read release policy - [X] I understand that unsupported versions don't get bug fixes. I will attempt to reproduce the issue on a supported version of Pulsar client and Pulsar broker. ### Version all released versions ### Minimal reproduce step Problem description: In the Shared subscription, messages get added to the replay queue when a consumer disconnects. In Key_Shared subscription, the replay queue is also used when messages cannot be dispatched to a target consumer due to insufficient permits or when the hash is blocked. There's a problem in the current implementation since the `dispatcherMaxReadSizeBytes` (default 5MB) setting isn't taken into account in the reads. The impact of this is that consumers will receive a large batch messages at once if the reads succeed. The exact implication of this isn't fully known at this time. However, it's against the design to ignore the `dispatcherMaxReadSizeBytes` setting which is helpful in making smaller incremental progress on individual dispatchers and service all active dispatchers in the broker one-by-one as fairly as possible. ### What did you expect to see? That `dispatcherMaxReadSizeBytes` is used for redelivering message. ### What did you see instead? `dispatcherMaxReadSizeBytes` is ignored. ### Anything else? _No response_ ### Are you willing to submit a PR? - [ ] I'm willing to submit a PR! -- 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]
