@zentol I think the decision to change javadocs or fix checkpoint coordinator's 
behavior depends on whether we give an explicit definition of the minimal pause 
between checkpoints. From our docs about [Minimum Pause Between 
Checkpoints](https://ci.apache.org/projects/flink/flink-docs-release-1.6/monitoring/checkpoint_monitoring.html#configuration-tab),
 it only said the successful scenario:

> After a checkpoint has completed successfully, we wait at least for this 
> amount of time before triggering the next one.

If we could give more clear definition on the checkpoint-failure scenario, it 
might not be a problem to discuss. 

[ Full content available at: https://github.com/apache/flink/pull/6346 ]
This message was relayed via gitbox.apache.org for [email protected]

Reply via email to