Github user kayousterhout commented on a diff in the pull request: https://github.com/apache/spark/pull/16867#discussion_r104517879 --- Diff: core/src/main/scala/org/apache/spark/scheduler/TaskSetManager.scala --- @@ -740,6 +743,7 @@ private[spark] class TaskSetManager( } removeRunningTask(tid) info.markFinished(state) + successfulTaskDurations.remove(taskInfos(tid).duration) --- End diff -- Ah I see you're saying that the problem is that the check that enough tasks have finished to enable speculation is no longer correct? If that's the sole issue, why don't we just keep using `tasksSuccessful` for that check rather than looking at the count of items in the heap?
--- 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