Github user vanzin commented on the pull request:
https://github.com/apache/spark/pull/8007#issuecomment-133199438
> 1) In my case, it was still YARN killing the container
I should have been more specific; in the preemption case, it's the RM
deciding whether to kill a container. So before the container is killed, the RM
already knows it's going away, and can potentially report that. I don't know
whether that's actually the case, which is why I wouldn't rely on that being
true.
In the overhead case, it's the NM killing the container; before the AM is
notified of that, the NM has to notify the RM. So there's plenty of room for
races in that case.
---
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]
