holdenk commented on a change in pull request #29014:
URL: https://github.com/apache/spark/pull/29014#discussion_r458376450
##########
File path:
core/src/main/scala/org/apache/spark/executor/CoarseGrainedExecutorBackend.scala
##########
@@ -166,11 +166,15 @@ private[spark] class CoarseGrainedExecutorBackend(
exitExecutor(1, "Received LaunchTask command but executor was null")
} else {
if (decommissioned) {
- logError("Asked to launch a task while decommissioned.")
+ val msg = "Asked to launch a task while decommissioned."
+ logError(msg)
Review comment:
I feel like I asked about this before, so sorry if it's a duplicate. Why
are we changing the messages to be on seperate lines from the log statement?
##########
File path:
core/src/main/scala/org/apache/spark/scheduler/TaskSchedulerImpl.scala
##########
@@ -912,13 +914,39 @@ private[spark] class TaskSchedulerImpl(
}
}
- override def executorDecommission(executorId: String): Unit = {
+ override def executorDecommission(
+ executorId: String, decommissionInfo: ExecutorDecommissionInfo): Unit = {
+ synchronized {
Review comment:
So my concern here is that if a task is scheduled part of the way
through decommissioning the executor *could* send back a decommission message
with the host lost. Maybe use a getOrElse to make sure if we've ever learnt the
entire host is going away we don't forget that?
----------------------------------------------------------------
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
