StephanEwen edited a comment on issue #9132: [FLINK-13245][network] Fix the bug of file resource leak while canceling partition request URL: https://github.com/apache/flink/pull/9132#issuecomment-514705841 I am wondering why we distinguish cases where we release with/without consumption notification. To me, all release should include a consumption notification - similar to what @azagrebin raised in the JIRA discussion. The way I understand it, `notifySubpartitionConsumed()` has a semantic of "a consumption attempt finished", irrespective of whether it was complete, incomplete, cancelled, etc. All release cases indicate that end of a consumption attempt. I understand that this is not exactly the behavior we had before, but if the above is true, the behavior before was arguably incorrect and we should fix this now.
---------------------------------------------------------------- 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] With regards, Apache Git Services
