You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Zhan Zhang (JIRA)" <ji...@apache.org> on 2014/10/27 22:55:34 UTC

[jira] [Commented] (SPARK-4103) Clean up SessionState in HiveContext

    [ https://issues.apache.org/jira/browse/SPARK-4103?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14185889#comment-14185889 ] 

Zhan Zhang commented on SPARK-4103:
-----------------------------------

There are already some efforts  (Spark-4037) targeting to this problem. The PR sent soon will serve the reference purpose and tries to make the solution more solid hopefully.

> Clean up SessionState in HiveContext
> ------------------------------------
>
>                 Key: SPARK-4103
>                 URL: https://issues.apache.org/jira/browse/SPARK-4103
>             Project: Spark
>          Issue Type: Bug
>          Components: SQL
>            Reporter: Zhan Zhang
>   Original Estimate: 12h
>  Remaining Estimate: 12h
>
> Currently the sessionState in HiveContext is lazy, and to start the session, SessionState.start() has to be invoked to initialize it. Using ThreadLocal variable natively in hive to track the availability of SessionState and initialize it if required. It avoid multiple thread share the same SessionState and single thread invoke SessionState.start with the same SessionState multiple times, which seems to cause unexpected behavior. 



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org