Github user vijikarthi commented on the pull request:
https://github.com/apache/incubator-zeppelin/pull/795#issuecomment-214963331
@Leemoonsoo The latest commits seems to be working. I could see two
different SC getting created in "isolated" mode. However when the second
SparkContext is getting created, I am hitting an exception which is related to
"SPARK-10872"
Regarding proxy user support, I am not sure if we can handle it inside
"SparkInterpreter". The way spark-submit handles proxy-user is by running the
main process itself as the proxy user (see
https://github.com/apache/spark/blob/master/core/src/main/scala/org/apache/spark/deploy/SparkSubmit.scala#L160).
In our case if we run the "RemoteInterpreterServer" as the proxy user, the
rest of the flow is kind of taken care automatically. I have tested
with/without Kerberos (by adding SPARK_SUBMIT_OPTIONS="--proxy-user vijay -v"
in bin/zeppelin-env.sh) and it works fine.
---
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.
---
