tillrohrmann commented on pull request #14847: URL: https://github.com/apache/flink/pull/14847#issuecomment-785303451
I think I have given bad advice on the savepoint clean up @XComp. I think that the `CheckpointCoordinator` will use the savepoint for recovery once it has been completed. Hence, we actually cannot delete the savepoint if a failure happens after the successful creation of the savepoint. Sorry for misguiding you here. The only solution I can think of atm is that we could log the path to the savepoint so that the user can manually clean it up after the job has been terminated. I think we can only properly solve this problem after [FLIP-47](https://cwiki.apache.org/confluence/display/FLINK/FLIP-47%3A+Checkpoints+vs.+Savepoints) has been implemented. ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: [email protected]
