[ 
https://issues.apache.org/jira/browse/SPARK-14003?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15670650#comment-15670650
 ] 

Herman van Hovell commented on SPARK-14003:
-------------------------------------------

This is a know issue. We use a single Hive client for all sessions within a 
SparkContext. 

> Multi-session can not work when one session is moving files for "INSERT ... 
> SELECT" clause
> ------------------------------------------------------------------------------------------
>
>                 Key: SPARK-14003
>                 URL: https://issues.apache.org/jira/browse/SPARK-14003
>             Project: Spark
>          Issue Type: Bug
>          Components: SQL
>    Affects Versions: 1.6.1
>            Reporter: Weizhong
>
> 1. Start ThriftServer
> 2. beeline A connect to ThriftServer, run "INSERT INTO ... SELECT"
> 3. when beeline A job is finished and start moving files, beeline B connect 
> to ThriftServer, then it will pending
> This is because we run client.loadDynamicPatitions in withHiveState, so it 
> will locked, then beeline B can't get the lock until beeline A finished 
> moving files.



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

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

Reply via email to