jiangxb1987 commented on a change in pull request #24497:
[SPARK-27630][CORE]Stage retry causes totalRunningTasks calculation to be
negative
URL: https://github.com/apache/spark/pull/24497#discussion_r284492288
##########
File path: core/src/main/scala/org/apache/spark/ExecutorAllocationManager.scala
##########
@@ -757,8 +753,11 @@ private[spark] class ExecutorAllocationManager(
val taskIndex = taskEnd.taskInfo.index
val stageId = taskEnd.stageId
allocationManager.synchronized {
- if (stageIdToNumRunningTask.contains(stageId)) {
- stageIdToNumRunningTask(stageId) -= 1
+ if (taskIdToNumLiveTasks.contains(taskId)) {
+ taskIdToNumLiveTasks(taskId) -= 1
+ if (taskIdToNumLiveTasks(taskId) <= 0) {
Review comment:
In that case if we don't expect negative number of tasks to happen I would
insert a assertion here, otherwise we can fail fast and print sufficient
information.
----------------------------------------------------------------
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]
