srowen commented on a change in pull request #22015:
[SPARK-20286][SPARK-24786][Core][DynamicAllocation] Release executors on
unpersisting RDD
URL: https://github.com/apache/spark/pull/22015#discussion_r258183175
##########
File path: core/src/main/scala/org/apache/spark/ExecutorAllocationManager.scala
##########
@@ -804,6 +824,14 @@ private[spark] class ExecutorAllocationManager(
}
}
+ override def onUnpersistRDD(unpersistRDD: SparkListenerUnpersistRDD): Unit
= {
Review comment:
Rather than trigger this extra check for idle executors and plumb through
additional fields, how about just checking cached blocks when onExecutorIdle is
already called? or is the point that it will never be called if its cache
timeout is set to a very long time?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on 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]
