Github user squito commented on a diff in the pull request:
https://github.com/apache/spark/pull/8760#discussion_r40225407
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/TaskSetManager.scala ---
@@ -58,15 +58,6 @@ private[spark] class TaskSetManager(
val conf = sched.sc.conf
- /*
- * Sometimes if an executor is dead or in an otherwise invalid state,
the driver
- * does not realize right away leading to repeated task failures. If
enabled,
- * this temporarily prevents a task from re-launching on an executor
where
- * it just failed.
- */
- private val EXECUTOR_TASK_BLACKLIST_TIMEOUT =
- conf.getLong("spark.scheduler.executorTaskBlacklistTime", 0L)
--- End diff --
we need to be sure to still support this conf name. In fact I think you
can just stick w/ this instead of switching to
"spark.scheduler.blacklist.timeout", right? perhaps that conf should only be
for strategies which do per-task blacklisting, but I feel like it could be for
all cases.
(In any case, I think we defer these exact naming decisions till we're
closer on the implementation, just mentioning it now while I'm noticing it.)
---
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 [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
