You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Thomas Graves (JIRA)" <ji...@apache.org> on 2014/04/14 18:28:16 UTC

[jira] [Created] (SPARK-1490) Add kerberos support to the HistoryServer

Thomas Graves created SPARK-1490:
------------------------------------

             Summary: Add kerberos support to the HistoryServer
                 Key: SPARK-1490
                 URL: https://issues.apache.org/jira/browse/SPARK-1490
             Project: Spark
          Issue Type: Improvement
          Components: Web UI
    Affects Versions: 1.0.0
            Reporter: Thomas Graves


Now that we have a history server that works on yarn and mesos we should add the ability for it to authenticate via kerberos so that it can read HDFS files without having to be restarted every 24 hours. 

One solution to this is to have the history server read a keytab file.  The Hadoop UserGroupInformation class has that functionality built in and as long as its using rpc to talk to hdfs it will automatically relogin when it needs to.   If the history server isn't using rpc to talk to hdfs then we would have to add some functionality to relogin approximately every 24 hours (configurable time).



--
This message was sent by Atlassian JIRA
(v6.2#6252)