gaoyunhaii commented on pull request #16714:
URL: https://github.com/apache/flink/pull/16714#issuecomment-896610683


   Very thanks Yuan @curcur for the review! 
   
   > I am fine with the current change. But maybe put a TODO in place?
   
   No problem, I added a TODO for this issue.
   
   > What would be the problem if a broadcast task with an empty state? Would 
it be possible that the task get more inputs as its upstream task has already 
sent finish without changing parallelism?
   
   Logically since broadcast is all_to_all, when part of the tasks containing 
broadcast are finished, the upstream tasks should all be finished. Thus after 
failover, there should be no more records (even if users change the topology, 
the topology change check should still keep the rule). Thus currently we may 
not have a specific case that would have problem due to this issue, but 
semantically we may first fix this issue so that it would avoid the future 
changes. 
   


-- 
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]


Reply via email to