Github user kayousterhout commented on a diff in the pull request: https://github.com/apache/spark/pull/15249#discussion_r81400344 --- Diff: core/src/main/scala/org/apache/spark/scheduler/TaskSetManager.scala --- @@ -592,34 +589,54 @@ private[spark] class TaskSetManager( * failures (this is because the method picks on unscheduled task, and then iterates through each * executor until it finds one that the task hasn't failed on already). */ - private[scheduler] def abortIfCompletelyBlacklisted(executors: Iterable[String]): Unit = { - - val pendingTask: Option[Int] = { - // usually this will just take the last pending task, but because of the lazy removal - // from each list, we may need to go deeper in the list. We poll from the end because - // failed tasks are put back at the end of allPendingTasks, so we're more likely to find - // an unschedulable task this way. - val indexOffset = allPendingTasks.lastIndexWhere { indexInTaskSet => - copiesRunning(indexInTaskSet) == 0 && !successful(indexInTaskSet) - } - if (indexOffset == -1) { - None - } else { - Some(allPendingTasks(indexOffset)) - } - } + private[scheduler] def abortIfCompletelyBlacklisted( + hostToExecutors: HashMap[String, HashSet[String]]): Unit = { + taskSetBlacklistOpt.foreach { taskSetBlacklist => + // If no executors have registered yet, don't abort the stage, just wait. We probably + // got here because a task set was added before the executors registered. + if (hostToExecutors.nonEmpty) { + // take any task that needs to be scheduled, and see if we can find some executor it *could* + // run on + val pendingTask: Option[Int] = { + // usually this will just take the last pending task, but because of the lazy removal + // from each list, we may need to go deeper in the list. We poll from the end because + // failed tasks are put back at the end of allPendingTasks, so we're more likely to find + // an unschedulable task this way. + val indexOffset = allPendingTasks.lastIndexWhere { indexInTaskSet => + copiesRunning(indexInTaskSet) == 0 && !successful(indexInTaskSet) + } + if (indexOffset == -1) { + None + } else { + Some(allPendingTasks(indexOffset)) + } + } - // If no executors have registered yet, don't abort the stage, just wait. We probably - // got here because a task set was added before the executors registered. - if (executors.nonEmpty) { - // take any task that needs to be scheduled, and see if we can find some executor it *could* - // run on - pendingTask.foreach { taskId => - if (executors.forall(executorIsBlacklisted(_, taskId))) { - val execs = executors.toIndexedSeq.sorted.mkString("(", ",", ")") - val partition = tasks(taskId).partitionId - abort(s"Aborting ${taskSet} because task $taskId (partition $partition)" + - s" has already failed on executors $execs, and no other executors are available.") + pendingTask.foreach { indexInTaskSet => + // try to find some executor this task can run on. Its possible that some *other* + // task isn't schedulable anywhere, but we will discover that in some later call, + // when that unschedulable task is the last task remaining. + val blacklistedEverywhere = hostToExecutors.forall { case (host, execs) => --- End diff -- execsOnHost? (o/w confusing on line 628)
--- 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