Github user CodingCat commented on a diff in the pull request:
https://github.com/apache/spark/pull/1313#discussion_r15518169
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/TaskSetManager.scala ---
@@ -751,20 +787,7 @@ private[spark] class TaskSetManager(
levels.toArray
}
- // Re-compute pendingTasksWithNoPrefs since new preferred locations may
become available
def executorAdded() {
- def newLocAvail(index: Int): Boolean = {
- for (loc <- tasks(index).preferredLocations) {
- if (sched.hasExecutorsAliveOnHost(loc.host) ||
- (sched.getRackForHost(loc.host).isDefined &&
- sched.hasHostAliveOnRack(sched.getRackForHost(loc.host).get))) {
- return true
- }
- }
- false
- }
- logInfo("Re-computing pending task lists.")
- pendingTasksWithNoPrefs =
pendingTasksWithNoPrefs.filter(!newLocAvail(_))
--- End diff --
en, because in current code, we only contain real "NO_PREF" tasks in the
pendingTasksWithNoPrefs, instead of also containing tasks whose preferences are
not available, see changes here:
https://github.com/apache/spark/pull/1313/files#diff-bad3987c83bd22d46416d3dd9d208e76L199,
so we don't recompute pendingTasksWithNoPrefs every time when the new
executor appears
---
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.
---
