hadoop-hive-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "John Sichi (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HIVE-1335) DataNucleus should use connection pooling
Date Thu, 06 May 2010 02:59:48 GMT

    [ https://issues.apache.org/jira/browse/HIVE-1335?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12864629#action_12864629
] 

John Sichi commented on HIVE-1335:
----------------------------------

If you use these settings, you'll get exactly what you're after:

{noformat}
<ivy-module version="2.0">
    <info organisation="org.apache.hadoop.hive" module="metastore"/>
    <dependencies>
        <dependency org="hadoop" name="core" rev="${hadoop.version.ant-internal}">
          <artifact name="hadoop" type="source" ext="tar.gz"/>
        </dependency> 
        <dependency org="commons-dbcp" name="commons-dbcp" rev="1.2.2">
          <exclude module="commons-pool" />
        </dependency>
        <dependency org="commons-pool" name="commons-pool" rev="1.2">
          <exclude module="commons-collections" />
          <exclude module="xerces" />
          <exclude module="xml-apis" />
        </dependency>
        <dependency org="org.datanucleus" name="datanucleus-connectionpool" rev="1.0.2">
          <exclude module="proxool" />
          <exclude module="c3p0" />
          <exclude module="datanucleus-core" />
          <exclude module="datanucleus-rdbms" />
          <exclude module="commons-collections" />
          <exclude module="commons-pool" />
        </dependency>
    </dependencies>
</ivy-module>

jsichi-mac:metastore jsichi$ ant ivy-retrieve
Buildfile: build.xml

ivy-init-dirs:

ivy-download:
      [get] Getting: http://repo2.maven.org/maven2/org/apache/ivy/ivy/2.1.0/ivy-2.1.0.jar
      [get] To: /Users/jsichi/open/hive-trunk/build/ivy/lib/ivy-2.1.0.jar
      [get] Not modified - so not downloaded

ivy-probe-antlib:

ivy-init-antlib:

ivy-init:

ivy-resolve:
[ivy:resolve] :: Ivy 2.1.0 - 20090925235825 :: http://ant.apache.org/ivy/ ::
[ivy:resolve] :: loading settings :: file = /Users/jsichi/open/hive-trunk/ivy/ivysettings.xml
[ivy:resolve] downloading http://repo1.maven.org/maven2/commons-dbcp/commons-dbcp/1.2.2/commons-dbcp-1.2.2.jar
...
[ivy:resolve] .........................................................
[ivy:resolve] ............................. (118kB)
[ivy:resolve] .. (0kB)
[ivy:resolve] 	[SUCCESSFUL ] commons-dbcp#commons-dbcp;1.2.2!commons-dbcp.jar (3271ms)
[ivy:resolve] downloading http://repo1.maven.org/maven2/commons-pool/commons-pool/1.2/commons-pool-1.2.jar
...
[ivy:resolve] ............................... (41kB)
[ivy:resolve] .. (0kB)
[ivy:resolve] 	[SUCCESSFUL ] commons-pool#commons-pool;1.2!commons-pool.jar (2412ms)
[ivy:resolve] downloading http://repo1.maven.org/maven2/org/datanucleus/datanucleus-connectionpool/1.0.2/datanucleus-connectionpool-1.0.2.jar
...
[ivy:resolve] ............ (14kB)
[ivy:resolve] .. (0kB)
[ivy:resolve] 	[SUCCESSFUL ] org.datanucleus#datanucleus-connectionpool;1.0.2!datanucleus-connectionpool.jar
(2683ms)

ivy-retrieve:

BUILD SUCCESSFUL
{noformat}

Personally, I agree that the cost/benefit of ivy is shaky, but if Hive as a project has chosen
it, then we should use it where possible.  If that decision needs to be revisited, then it
would be good to resolve that.


> DataNucleus should use connection pooling
> -----------------------------------------
>
>                 Key: HIVE-1335
>                 URL: https://issues.apache.org/jira/browse/HIVE-1335
>             Project: Hadoop Hive
>          Issue Type: Improvement
>            Reporter: Edward Capriolo
>            Assignee: Edward Capriolo
>             Fix For: 0.6.0
>
>         Attachments: commons-dbcp-1.2.2.jar, commons-dbcp.LICENSE, commons-pool-1.2.jar,
commons-pool.LICENSE, datanucleus-connectionpool-1.0.2.jar, datanucleus-connectionpool.LICENSE,
hive-1335.patch.txt
>
>
> Currently each Data Nucleus operation disconnects and reconnects to the MetaStore over
jdbc. Queries fail to even explain properly in cases where a table has many partitions. This
is fixed by enabling one parameter and including several jars.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message