Github user vanzin commented on the pull request:
https://github.com/apache/spark/pull/8668#issuecomment-139306000
@KaiXinXiaoLei I still don't understand what's the issue.
- Executors are hung / dead / in gc hell and don't send heartbeats
- Hearbeat receiver kills them. All traces of the dead executor (including
any idle timeouts) are lost. Code asks for new executors to replace them.
- New executors come up, register, have fresh idle timeouts.
So, aside from the new executors living for a little longer than the old
ones would, what is the issue? Where is the count wrong?
---
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]
