LantaoJin commented on issue #25971: [SPARK-29298][CORE] Separate block manager heartbeat endpoint from driver endpoint URL: https://github.com/apache/spark/pull/25971#issuecomment-548630364 https://github.com/apache/spark/blob/0da667d31436c43e06cb6bb5ac65a17f65edd08b/core/src/main/scala/org/apache/spark/scheduler/DAGScheduler.scala#issuecomment-548545864 Yes. In a heavy cluster, it will timeout and cause block manager re-register unexpected. So that’s what the PR to fix, not just increase the timeout value. This code still keep `syncAsk` with 10 minutes timeout. Instead, the response from new heartbeat endpoint is fast even when the driver is heavy since the logic in heartbeat endpoint is light.
---------------------------------------------------------------- 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]
