[
https://issues.apache.org/jira/browse/SPARK-15940?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15329292#comment-15329292
]
Sean Owen commented on SPARK-15940:
-----------------------------------
Multiple contexts in one process are not supported. Is that your use case?
> Simultaneous multiple spark context with different kerberos FID credentials
> from within one id
> ----------------------------------------------------------------------------------------------
>
> Key: SPARK-15940
> URL: https://issues.apache.org/jira/browse/SPARK-15940
> Project: Spark
> Issue Type: Bug
> Reporter: Partha Pratim Ghosh
>
> Spark authorization can be brought about by kerberos authentication. Now if
> we want to have two groups of users working on Spark contexts authorized by
> different ids, how to go about doing that?
> Our J2EE application is running as user1. From there we are opening 2 Spark
> contexts and we need to open them as follows -
> Spark Context (SC1) - kinit Functional Id (FID1)
> Spark Context (SC2) - kinit Functional Id (FID2)
> We can open the Spark contexts one by one. But after the Spark contexts are
> up we are using them, in many cases, parallelly. So, there are scenarios
> where we are firing SC1 and SC2 jobs parallelly. In that case, if the default
> kinit (the last kinit that was done) is FID2 and we run a job on SC1 then we
> get exception as FID1 does not get validated.
> How do we maintain 2 spark contexts through one single user (user1), getting
> authorized by 2 different FIDs? We need this functionality as the
> authorization differs. Please advise.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
