Github user GayathriMurali commented on a diff in the pull request:
https://github.com/apache/spark/pull/11544#discussion_r55908239
--- Diff:
core/src/main/scala/org/apache/spark/ui/jobs/JobProgressListener.scala ---
@@ -374,28 +374,30 @@ class JobProgressListener(conf: SparkConf) extends
SparkListener with Logging {
execSummary.taskTime += info.duration
stageData.numActiveTasks -= 1
- val (errorMessage, metrics): (Option[String], Option[TaskMetrics]) =
+ val (errorMessage, metrics): (Option[String], Seq[AccumulableInfo]) =
taskEnd.reason match {
case org.apache.spark.Success =>
stageData.completedIndices.add(info.index)
stageData.numCompleteTasks += 1
- (None, Option(taskEnd.taskMetrics))
- case e: ExceptionFailure => // Handle ExceptionFailure because
we might have metrics
+ (None, taskEnd.taskMetrics.accumulatorUpdates())
+ case e: ExceptionFailure => // Handle ExceptionFailure because
we might have metrics
stageData.numFailedTasks += 1
- (Some(e.toErrorString), e.metrics)
- case e: TaskFailedReason => // All other failure cases
+ (Some(e.toErrorString), e.accumUpdates)
+ case e: TaskFailedReason => // All other failure cases
stageData.numFailedTasks += 1
- (Some(e.toErrorString), None)
+ (Some(e.toErrorString), Seq.empty[AccumulableInfo])
}
- metrics.foreach { m =>
+ val taskMetrics: Option[TaskMetrics] = if (metrics.nonEmpty)
Option(TaskMetrics.
+ fromAccumulatorUpdates(metrics)) else None
--- End diff --
Can you please explain the idea behind using Some instead of Option? When
`fromAccumulatorUpdates` is enclosed within a if that checks if metrics is
nonempty, we are in the safe zone.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
