315157973 commented on pull request #11325: URL: https://github.com/apache/pulsar/pull/11325#issuecomment-938602811
> > The current reading process is like this: > > > > 1. Determine whether the threshold is exceeded > > 2. Enter the pending queue and prepare to read > > 3. The entry is successfully read, and the threshold count is deducted > > > > There is a time difference between 1 and 3. When the Broker just restarts, if it is read concurrently, the problem in that PR will still exists > > The problem you mentioned may be solved in this PR #8611 This PR will only prevent the message from being sent to the consumer for a long time later, but it cannot solve the memory surge during concurrent reading when broker restart. -- 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]
