SEZ9 commented on issue #9940: URL: https://github.com/apache/seatunnel/issues/9940#issuecomment-5691301015
@BarryQian, to recap where this stands: #10778 (merged into `dev`) fixes the checkpoint restore / failover path where restored RocketMQ offsets were overwritten and consumption restarted from offset 0. A cold start with `commit.on.checkpoint = false` and no broker-side committed offset for the consumer group follows RocketMQ group-offset behavior and is not a SeaTunnel bug. If you still see consumption restarting from offset 0 on a build that contains #10778, could you share: 1. the restart type (plain start vs. checkpoint / failover restore) and the `commit.on.checkpoint` value used, 2. whether the broker shows committed offsets for `track_report_group` at that moment, and 3. any relevant warnings from the job log around startup. Also, the config above sets `format` twice; please confirm the config actually used. If the checkpoint-restore duplicate consumption no longer reproduces on `dev`, we can close this issue. <!-- streview-comment:1079 --> -- 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]
