Github user ueshin commented on the pull request:
https://github.com/apache/spark/pull/6352#issuecomment-104647426
@srowen, Thank you for your checking.
1. To check the parent stages are missing or not, we only have to check the
location of RDD that has `ShuffleDependency` because narrow-depending RDDs
would be the same location, so I moved the calling `getCacheLocs` into the
`case shufDep`.
2. I was afraid that the dependency graph would become bigger (if there are
a lot of `union` or `zipPartitions`, etc.) so I added checking the location at
the point the RDD has more than 1 dependencies.
* we might not need [line
389](https://github.com/apache/spark/pull/6352/files#diff-6a9ff7fb74fd490a50462d45db2d5e11R389)
if we don't have to consider the case.
---
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]
