jiangxb1987 commented on issue #25971: [SPARK-29298][CORE] Separate block manager heartbeat endpoint from driver endpoint URL: https://github.com/apache/spark/pull/25971#issuecomment-552010809 > so other events do not have timeout? or they will retry if timeout? Please correct me if I'm wrong but I don't see approach to retry when `GetLocations*` requests timeout. > It's still useful to have a busy block manager as other events can handle timeout gracefully. If this is the case then the approach proposed by this PR is fine, otherwise, all the requests towards the `BlockManagerMasterEndpoint` just timeout, we might need to figure out other ways to make things work. > Heartbeat could be asynchronous handled. I won't call an async message `Heartbeat`.
---------------------------------------------------------------- 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 --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
