Github user pwendell commented on the pull request:
https://github.com/apache/spark/pull/586#issuecomment-41714679
@tgravescs ah I see, you're right. I think I assumed incorrectly that the
executor launcher would bundle up the options and send them over, but I don't
actually see that happening anywhere. So this part of the code is actually not
used:
https://github.com/apache/spark/blob/df6d81425bf3b8830988288069f6863de873aee2/yarn/common/src/main/scala/org/apache/spark/deploy/yarn/ClientBase.scala#L328
What happens is the executor is just getting its configuration from the
driver when the executor launches. And that works in _most_ cases except for
security, which it needs to know about before connecting. Is that right?
---
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.
---
