azagrebin commented on pull request #14199: URL: https://github.com/apache/flink/pull/14199#issuecomment-734883917
I added TM failure for batch case. The JM failure is tricky w/o synchronization with the job execution. The job gets done before JM failure. We could make `EmbeddedHaServices::getJobManagerLeaderService` public to see whether the JM is still available to restart but this is fragile to concurrency. Hence, I am not sure about the value of this test. I added JM failure test for batch with the job execution synchronization, abstracting it away, in the last commit. I can remove the last commit if it looks too complicated and there is no better idea how to do the JM failure test for batch in a simpler way. ---------------------------------------------------------------- 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]
