Github user srowen commented on the pull request:
https://github.com/apache/spark/pull/5550#issuecomment-93954322
OK, the screenshot and new description clarify it. Stage 0 completes, so is
counted towards the total and completed stage IDs. It is retried because Stage
1 failed (?) and is skipped. It's still considered to have completed, but now
does not count against the total since it was also skipped (the second time).
The same logic works for tasks since tasks are not retried under the same
ID, I think?
The problem with this change as it is, is what I mentioned -- if a stage
*doesn't* complete and is skipped then you have something like "2/3 (1
skipped)" which isn't great.
However to fix this, should a stage's ID be removed from the set of
completed stages if it is retried?
CC @JoshRosen
---
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]
