Github user tgravescs commented on the issue:
https://github.com/apache/spark/pull/13482
So why don't we just take out the notifyAll call when we get a
GetExecutorLossReason?
We can add a parameter to resetAllocatorInterval() and resets the interval
it but doesn't call the notify. The most its going to sleep is the heartbeat
interval (a few seconds), which doesn't seem bad to just get the lost reason.
If they keep coming in the interval gets reset so it only sleep 200ms (default)
on subsequent sleeps.
I guess you might end up with the same situation on the RequestExecutors,
but generally there I expect we request a bunch at a time and the allocation
manager runs on a timer so those shouldn't be happens that frequently.
And we probably don't really need the notify on the requestExecutors call
either but it its not hurting anything it could speed up requests a few seconds.
---
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]
