Mailing-List: contact dev-help@hive.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@hive.apache.org
Date: Wed, 20 Feb 2013 18:43:12 +0000 (UTC)
From: "Andrew Tindle (JIRA)" <jira@apache.org>
To: hive-dev@hadoop.apache.org
Message-ID: <JIRA.12633250.1361385752421.311571.1361385792738@arcas>
In-Reply-To: <JIRA.12633250.1361385752421@arcas>
References: <JIRA.12633250.1361385752421@arcas>
Subject: [jira] [Created] (HIVE-4043) Parallel Hive Queries: Sporadic Errors
 of form: Error in metadata: javax.jdo.JDOFatalDataStoreException: IO Error:
 Connection reset
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit

Andrew Tindle created HIVE-4043:
-----------------------------------

             Summary: Parallel Hive Queries: Sporadic Errors of form: Error in metadata: javax.jdo.JDOFatalDataStoreException: IO Error: Connection reset
                 Key: HIVE-4043
                 URL: https://issues.apache.org/jira/browse/HIVE-4043
             Project: Hive
          Issue Type: Bug
          Components: Metastore
    Affects Versions: 0.9.0
         Environment: O/S: RHEL 6.3
Metastore: Oracle 11gR2
            Reporter: Andrew Tindle


I have a program that spawns Hive queries/processes, up to a maximum of 5, in parallel. When the number of queries drops below. ie the process has ended, another Hive query/process is initiated.

Sometimes, this program works, i.e. all 34 queries successfully process.

However, on other occasions, I get sporadic instances of the following error for some of the queries:

FAILED: Error in metadata: javax.jdo.JDOFatalDataStoreException: IO Error: Connection reset
NestedThrowables:
java.sql.SQLRecoverableException: IO Error: Connection reset
FAILED: Execution Error, return code 1 from org.apache.hadoop.hive.ql.exec.DDLTask

Can anyone help in identifying/resolving why this occurs. It looks to me as if there is some kind of race condition/collision with the Hive Metastore, this being hosted in an Oracle DB on the same node as the Hadoop infrastructure (single node).


--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira