[GitHub] azagrebin edited a comment on issue #7186: [FLINK-10941] Keep slots which contain unconsumed result partitions

[GitBox <gi...@apache.org>]

azagrebin edited a comment on issue #7186: [FLINK-10941] Keep slots which contain unconsumed result partitions
URL: https://github.com/apache/flink/pull/7186#issuecomment-468340433
 
 
   Thanks for working on this @QiLuo-BD @zhijiangW 
   
   I have a question about `CloseRequest` reliability. What if this message does not reach the producer? E.g. dropped by network. Or will this kind of failure properly fail consumer and be propagated to JobManager to properly release partitions in producer.
   
   It seems to hold the release of `TaskExecutor`. Currently, it seems to lead to resources leak only on the `TaskExecutor` while is still running.
   
   I am mot sure about current yarn deployment but if the `TaskExecutor`'s are not reused there (rather always newly allocated), it will lead to growing number of idle executors in case of abrupt job failures.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services