vanzin commented on a change in pull request #23698: [SPARK-26766][CORE] Remove
the list of filesystems from
HadoopDelegationTokenProvider.obtainDelegationTokens
URL: https://github.com/apache/spark/pull/23698#discussion_r254380466
##########
File path:
core/src/main/scala/org/apache/spark/deploy/security/HadoopFSDelegationTokenProvider.scala
##########
@@ -133,3 +133,45 @@ private[deploy] class HadoopFSDelegationTokenProvider
if (renewIntervals.isEmpty) None else Some(renewIntervals.min)
}
}
+
+private[deploy] object HadoopFSDelegationTokenProvider {
+ def hadoopFSsToAccess(
+ sparkConf: SparkConf,
+ hadoopConf: Configuration): Set[FileSystem] = {
+ val filesystemsToAccess = sparkConf.get(KERBEROS_FILESYSTEMS_TO_ACCESS)
+
+ val defaultFS = FileSystem.get(hadoopConf)
+ val master = sparkConf.get("spark.master", null)
+ val stagingFS = if (master != null && master.contains("yarn")) {
+ sparkConf.get(STAGING_DIR)
+ .map(new Path(_).getFileSystem(hadoopConf))
+ .getOrElse(defaultFS)
Review comment:
You're already adding `defaultFS` explicitly to the final list. So if
`STAGING_DIR` is not defined, you'll have `defaultFS` twice. It's ok because
the final list is a set, but why would you do that?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
