Github user mgummelt commented on a diff in the pull request:
https://github.com/apache/spark/pull/18098#discussion_r118586090
--- Diff:
resource-managers/mesos/src/main/scala/org/apache/spark/scheduler/cluster/mesos/MesosCoarseGrainedSchedulerBackend.scala
---
@@ -502,6 +521,25 @@ private[spark] class
MesosCoarseGrainedSchedulerBackend(
)
}
+ private def satisfiesLocality(offerHostname: String): Boolean = {
+ if (hostToLocalTaskCount.nonEmpty) {
--- End diff --
You've agreed that the semantics should be "Launch an executor on a host
only when we have a task that wants to be on that host, or the configurable
delay has elapsed"
By launching tasks on any arbitrary host when no locality info is
available, you're violating those semantics, because even before the delay has
elapsed, the scheduler will launch a task on an agent that no task wants to be
on. Does that make sense?
---
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
