holdenk commented on a change in pull request #29367:
URL: https://github.com/apache/spark/pull/29367#discussion_r468207413
##########
File path:
core/src/test/scala/org/apache/spark/scheduler/WorkerDecommissionSuite.scala
##########
@@ -76,7 +76,9 @@ class WorkerDecommissionSuite extends SparkFunSuite with
LocalSparkContext {
// decom.sh message passing is tested manually.
val sched = sc.schedulerBackend.asInstanceOf[StandaloneSchedulerBackend]
val execs = sched.getExecutorIds()
- execs.foreach(execId => sched.decommissionExecutor(execId,
ExecutorDecommissionInfo("", false)))
+ // Make the executors decommission, finish, exit, and not be replaced.
+ val execsAndDecomInfo = execs.map((_, ExecutorDecommissionInfo("", false)))
+ sched.decommissionExecutors(execsAndDecomInfo, adjustTargetNumExecutors =
true)
Review comment:
My reasoning here is that by adjusting the executors down here we ensure
the task isn't being re-launched and a re-launched executor so it improves the
test.
##########
File path:
core/src/test/scala/org/apache/spark/storage/BlockManagerDecommissionIntegrationSuite.scala
##########
@@ -190,7 +190,8 @@ class BlockManagerDecommissionIntegrationSuite extends
SparkFunSuite with LocalS
logInfo(s"Decommissioning executor ${execToDecommission}")
sched.decommissionExecutor(
execToDecommission,
- ExecutorDecommissionInfo("", isHostDecommissioned = false))
+ ExecutorDecommissionInfo("", isHostDecommissioned = false),
+ adjustTargetNumExecutors = true)
Review comment:
Same.
----------------------------------------------------------------
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]
