Github user steveloughran commented on the pull request:
https://github.com/apache/spark/pull/4491#issuecomment-83972651
Most of the sync changes in HADOOP-1170 was to deal with HBase's
expectations. Even if you don't think it matters for its current use in spark,
you should make the `close()` operation thread safe. It currently isn't.
Looking at hadoop trunk, recent changes to crypto packages include
HADOOP-11674 , and HADOOP-11343 ; the latter looks like it's diverged from your
code. I recommend you checkout hadoop/trunk locally , log recent changes to the
crypto packages & track new ones, as things are still stabilising
w.r.t YARN NM resolving library differences: if the NM is handling the
shuffle, it will have the relevant native lib for its hadoop-common.jar on its
path, so will be in sync. Though that doesn't address the problem for the
recipient, does it?
---
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]
