New method for obtaining report of NameNode and JobTracker internals
--------------------------------------------------------------------
Key: HADOOP-215
URL: http://issues.apache.org/jira/browse/HADOOP-215
Project: Hadoop
Type: New Feature
Components: mapred, dfs
Reporter: alan wootton
Priority: Minor
Many weeks ago we (at shopping.com research) decided we wanted to be able to
get reports from the internals of JobTracker and NameNode. The hadoop web
server provides some of this, but we wanted a more structured output, and
easier extensibility.
So, we decided to use xml, and I wrote it.
There is a very thin interface to ClientProtocol, and JobSubmissionProtocol
like this:
public XmlReporter getXmlReport(String classname, String question);
The implementation (in JobTracker and NameNode ) looks like this:
public XmlReporter getXmlReport(String classname, String question)
{
XmlReporter reporter = XmlReporter.getInstance( classname,
this, question);
reporter.report();
return reporter;
}
The idea being that you pass in some xml (question), an XmlReporter (classname)
is instanciated and passed back.
An XmlReporter object consists of nothing more that two org.w3c.dom.Document
objects (one in the question, the other is the answer). The Writable interface,
for the RPC, simply serializes the dom tree to a string, and then parses it
back to a dom tree.
Anyway, here it is. I would like for it to either make it into the code, or for
me to find anoher way.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see:
http://www.atlassian.com/software/jira
