Github user tgravescs commented on the pull request:
https://github.com/apache/spark/pull/607#issuecomment-41963593
I think you would want it even higher then just wrapping the thread pool.
Looking at the CoarseGrainExecutorBackend for instance I think you would have
it wrap the run routine. This way your akka connection gets setup with it as
the correct user. You also don't have to worry about someone changing the
executor to have something important out side of the wrapping of thread pool.
I'm not sure how this affect mesos though. Does mesos run things as the
user who launched, as a super user, or other? I assume its been working as it
is now so moving it up into the MesosExecutorBackend is ok. I'm just not sure
if it should wrap everything (like MesosExecutorDriver)
---
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.
---
