ans76 opened a new issue #6949: URL: https://github.com/apache/incubator-doris/issues/6949
### Search before asking - [X] I had searched in the [issues](https://github.com/apache/incubator-doris/issues?q=is%3Aissue) and found no similar issues. ### Version palo 0.14.7 ### What's Wrong? Query against historical snapshot table(without any write jobs), get unstable results from the multiple queries. these tablets are not in InconsistentTablets if using `SHOW PROC '/statistic/db_id';` check table healthy status, found suspicious info from each replica detail info. replica 154845761's `VersionCount`,`DataSize` and `RowCount` were different from other replicas but got same `LstSuccessVersion` and `LstFailedVersion` is -1, I guess that's the reason `TabletChecker` cannot identify these tablets unhealth, correct me if i wrong.  `ADMIN CHECK TABLET` will [calculate checksum](https://github.com/apache/incubator-doris/blob/23bd4a959ade68a6b1eafe462ceea64a5acc714d/be/src/olap/task/engine_checksum_task.cpp#L105-L115) based on each row, what if checksum of replicas were different and report it to fe, is there any procedure on fe will treat `inconsistent tablets` rather `unhealthy tablets`? ``` /** * A replica is healthy only if * 1. the backend is available * 2. replica version is caught up, and last failed version is -1 * * A tablet is healthy only if * 1. healthy replica num is equal to replicationNum * 2. all healthy replicas are in right cluster */ ``` ### What You Expected? - Get stable result from same query ### How to Reproduce? _No response_ ### Anything Else? _No response_ ### Are you willing to submit PR? - [ ] Yes I am willing to submit a PR! ### Code of Conduct - [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct) -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
