RongtongJin commented on issue #10666: URL: https://github.com/apache/rocketmq/issues/10666#issuecomment-5138781774
Thanks for the detailed report. This looks like an important Controller-mode HA data-integrity issue. I do not see a linked PR or anyone explicitly working on it yet. @chenxu80, would you be able to help submit a focused fix for this? The current proposed direction sounds reasonable to me: make the truncation durable by invalidating or zeroing the stale bytes after the truncation offset in the boundary `MappedFile`, and add regression coverage for the truncate + restart recovery scenario so stale CommitLog records are not recovered after the slave has truncated to the consistent point. If you plan to work on it, please leave a note here so we can coordinate review. -- 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]
