Github user rvesse commented on a diff in the pull request:
https://github.com/apache/spark/pull/21669#discussion_r208531808
--- Diff: core/src/main/scala/org/apache/spark/deploy/SparkSubmit.scala ---
@@ -164,7 +164,15 @@ private[spark] class SparkSubmit extends Logging {
}
}
} else {
- runMain(childArgs, childClasspath, sparkConf, childMainClass,
args.verbose)
+ if
(sparkConf.getOption("spark.kubernetes.kerberos.proxyUser").isDefined) {
+ // scalastyle:off println
+ printStream.println("Running as proxy user in k8s cluster
mode...")
+ // scalastyle:on println
+ SparkHadoopUtil.get.runAsSparkUser(
+ () => runMain(childArgs, childClasspath, sparkConf,
childMainClass, args.verbose))
--- End diff --
So if I have followed the code flow correctly now when using Kerberos auth
the value of `SPARK_USER` provided to the driver is the user we're launching
the job as from a Spark/HDFS perspective?
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
