tillrohrmann commented on pull request #16543: URL: https://github.com/apache/flink/pull/16543#issuecomment-884992135
In order to reproduce the problem you have to delay the registration of the `TaskExecutor` a bit. You could do this by adding `Thread.sleep(1000L)` to the `TaskExecutor.onStart()` method. That way you can reproduce the race condition between the `TaskExecutor` registration and the arrival of the slot request that would fail after the start up period is over. I hope this makes the problem a bit clearer. -- 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]
