Ngone51 commented on a change in pull request #22806: [SPARK-25250][CORE] :
Late zombie task completions handled correctly even before new taskset launched
URL: https://github.com/apache/spark/pull/22806#discussion_r247779264
##########
File path:
core/src/main/scala/org/apache/spark/scheduler/TaskSchedulerImpl.scala
##########
@@ -286,6 +286,44 @@ private[spark] class TaskSchedulerImpl(
}
}
+ /**
+ * SPARK-25250: Whenever any Task gets successfully completed, we simply
mark the
+ * corresponding partition id as completed in all attempts for that
particular stage and
+ * additionally, for a Result Stage, we also kill the remaining task
attempts running on the
+ * same partition. As a result, we do not see any Killed tasks due to
+ * TaskCommitDenied Exceptions showing up in the UI. When this method is
called from
+ * DAGScheduler.scala on a task completion event being fired, it is assumed
that the new
+ * TaskSet has already been created and registered. However, a small
possibility does exist
+ * that when this method gets called, possibly the new TaskSet might have
not been added
Review comment:
I think @squito has a good ponit here. Previously, I was thinking what if
the active TaskSet has not been created when we marking completed partition for
all TaskSets and does this fix still works ? Now, I realize that whether the
active TaskSet has been created or not, it still works:
* created
obviously, fine.
* not created
then, when `DAGScheduler` calling `submitMissingTasks`, it will figure out
which missing partitions to compute(including the partitions which were
completed by tasks from previous stage attempt). So, the new created TaskSet
also know about the completed partition. And these are all benefit from event
loop, which perform as a single thread.
----------------------------------------------------------------
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]
