Github user mateiz commented on the pull request:
https://github.com/apache/spark/pull/7699#issuecomment-135802458
Regarding the "wider class of problems", I just meant that the core problem
here seems to be that tasks don't get identified correctly. This also seems to
affect other code paths, e.g. if a task is Resubmitted. Would it be possible to
test for that right now without additional test infrastructure?
Otherwise, I think it's fine to merge this as is, but we definitely have to
"zombify" the old stage. If you want to work on that, it would be great if you
could write a doc that explains the state transitions and, in particular, shows
that the stage will always become a zombie if we submit another copy of it.
Otherwise it's hard to reason from just a patch.
---
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]
