Github user tdas commented on the issue:
https://github.com/apache/spark/pull/12576
Actually, correction. Here is a better fix. The key problem here is that
the RDD internal logic for computing from checkpoints, depends on a public,
override-able API. This is wrong. Either the `isCheckpointed` should have been
marked final, or the internal `isCheckpointedAndMaterialized` should not have
depended on `isCheckpointed` (see
[here](https://github.com/apache/spark/blob/master/core/src/main/scala/org/apache/spark/rdd/RDD.scala#L1592)).
So the simplest, and correct solution is probably (not tested) to change
`isCheckpointedAndMaterialzed`.
```
private[spark] def isCheckpointedAndMaterialized: Boolean =
checkpointData.exists(_.isCheckpointed)
```
```
---
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]
