redlsz opened a new issue, #7724: URL: https://github.com/apache/rocketmq/issues/7724
### Before Creating the Bug Report - [X] I found a bug, not just asking a question, which should be created in [GitHub Discussions](https://github.com/apache/rocketmq/discussions). - [X] I have searched the [GitHub Issues](https://github.com/apache/rocketmq/issues) and [GitHub Discussions](https://github.com/apache/rocketmq/discussions) of this repository and believe that this is not a duplicate. - [X] I have confirmed that this bug belongs to the current repository, not other repositories of RocketMQ. ### Runtime platform environment centos7 ### RocketMQ version 5.1.0 ### JDK Version JDK 1.8 ### Describe the Bug A message is popped up again after being acknowledged, even appears twice at the same time. No offset-reset or service-reboot. queryMsgByOffset results: <img width="628" alt="image" src="https://github.com/apache/rocketmq/assets/103550934/bfcb7d91-6a48-458d-9962-e36a46dc8fa5";> It can be noticed that broker store two identical messages (same UNIQ_KEY) to pop-retry topic at the same time. And this message has actually been acknowledged before. After reviewing the codes, we have a suspicion about the _reviveMsgFromCk_: When revive a message from a PopCheckPoint, the original message should be read out from store firstly. But what the _getBizMessage_ actual does is reading one message **starting from** the specified queue offset, instead of message **of** the specified queue offset. In other words, incorrect message may be read out and revived in some scenarios. (In addition, although the pop-time-check may avoids the above issue to some extent, we should ensure that a message's pop-time will never be less than its store-time.) <img width="1530" alt="image" src="https://github.com/apache/rocketmq/assets/103550934/2fbb1314-c649-4fe9-924e-3a6f8706fef0";> ### Steps to Reproduce A possible case: message in queue: m0(storeTime=t0,commitlogfile0), m1(storeTime=t1,commitlogfile1) timeline: t2 -> Pop m0 (donnot ack) t3 -> commitlogfile0 is expired and deleted t4 -> invisible timeout and try revive m0, but the actual read and revived messag is m1 (t0<t1<t2<t3<t4) ### What Did You Expect to See? Message will not be received again after being acknowledged. ### What Did You See Instead? Message is received again after being acknowledged. ### Additional Context _No response_ -- 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]
