@azagrebin thanks for your suggestion, I agree and reconsider more details.

Considering that @tillrohrmann  has said that the current checkpoint exception 
handler should also be implemented in `CheckpointCoordinator`. 

> I understand why you implemented it the way you did. I think the 
> `setFailOnCheckpointingErrors` should actually also be handled by the 
> `CheckpointCoordinator`/`JM` and not on the Task level (but this is a 
> different story). This looks a little bit like a shortcut we made back in the 
> days.

Do we need to take this refactoring into account? Because this PR is actually a 
supplement to the checkpoint exception handler.

[ Full content available at: https://github.com/apache/flink/pull/6567 ]
This message was relayed via gitbox.apache.org for [email protected]

Reply via email to