sodonnel commented on PR #4332: URL: https://github.com/apache/hadoop/pull/4332#issuecomment-1133193488
Yea, the BackoffMonitor already used the cancelled nodes technique. Initially, I tried to split the the stopMaintenance method, but then I realised there is a piace in the check loop where it yields the write lock, so someone could cancel a node and it would slip in while the lock is dropped, and could still cause the issue. So I figured it was better to do the asynchronous cancel in the same way as the backOff monitor to fix it completely. -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
