hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Doug Cutting (JIRA)" <j...@apache.org>
Subject [jira] Updated: (HADOOP-1518) Add session id to metric data
Date Thu, 21 Jun 2007 22:14:26 GMT

     [ https://issues.apache.org/jira/browse/HADOOP-1518?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel

Doug Cutting updated HADOOP-1518:

    Status: Open  (was: Patch Available)

The javadoc for set/getSessionId should describe what the session id is used for, as should
the description in hadoop-default.xml.  Will this actually be set in site configurations,
or only programatically?  If it's only set programatically, then it doesn't need to be in
hadoop-default.xml, which primarily exists to document available parameters for site configuration,
not necessarily every parameter ever used by programs. 

> Add session id to metric data
> -----------------------------
>                 Key: HADOOP-1518
>                 URL: https://issues.apache.org/jira/browse/HADOOP-1518
>             Project: Hadoop
>          Issue Type: Improvement
>            Reporter: David Bowen
>            Assignee: David Bowen
>         Attachments: patch.txt
> In the context of Hadoop-On-Demand (HOD), we want to be able to aggregate metrics across
the machines in the virtual cluster that corresponds to a HOD session. To enable this, we'd
like to tag all the map/reduce related metrics with a session id.  The session id would be
a string created by HOD (it might be a Torque job id) and put into the hadoop-site.xml file
for the virtual cluster.
> The default value for the session id would be the empty string, so that non-HOD usage
of the metrics is not affected accept for the addition of a tag which is always the empty

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

View raw message