Github user vanzin commented on a diff in the pull request:
https://github.com/apache/spark/pull/18616#discussion_r127314530
--- Diff:
core/src/main/scala/org/apache/spark/deploy/security/HadoopDelegationTokenManager.scala
---
@@ -42,7 +42,7 @@ import org.apache.spark.internal.Logging
private[spark] class HadoopDelegationTokenManager(
sparkConf: SparkConf,
hadoopConf: Configuration,
- fileSystems: Set[FileSystem])
+ fileSystems: () => Set[FileSystem])
--- End diff --
> the auth to visit FS is still "SIMPLE"
That sounds like your HDFS configuration is missing something. The
`yarnConf` object passed to `hadoopFSsToAccess()` should have Kerberos
configured as the auth mechanism, and because the container has valid
delegation tokens, the HDFS library should use them.
Otherwise, I don't see how this feature would work at all; when submitting
an app in cluster mode, the driver does not do a kerberos login (until tokens
need to be renewed), and yet applications are able to talk to HDFS.
---
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]
