89799969 commented on issue #10745: URL: https://github.com/apache/rocketmq/issues/10745#issuecomment-5564993431
Thanks for the detailed review. Your analysis is correct. I reproduced the exact pre-change one-sided close behavior with paired EmbeddedChannel pipelines and drained the pending event-loop tasks. In all four paths (frontend/backend write failure and frontend/backend exception), both channels became inactive through the existing channelInactive cascade: 4 tests run, 0 failures. This shows that the tests in #11052 asserted the proposed implementation detail rather than demonstrating a retained active channel. I am closing #11052 because the resource-leak premise is not supported. -- 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]
