The scheduling and resource management parts of the JobTracker have been
put in the ResourceManager.
The Map Reduce job history in the JobHistoryServer. Each MR job is managed
by its own YARN MR Application.
From: Tiago Kepe [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, December 18, 2012 2:02 AM
To: [EMAIL PROTECTED]
Subject: Re: How to compile the JobTracker.java
Thanks by answer.
Some thinks was clarified for me. I am learning about MR2, I realized that
in MR2 the JobTracker was replaced by JobHistory, is this correct?
Could you please provide more information or link about MR2?
2012/12/18 Harsh J <[EMAIL PROTECTED]>
> You seem to be using trunk to modify "JobTracker.java". However, MR1
> has been removed out of trunk in favor of MR2 applications, and hence
> the code you'll find for it is almost of no use and is unmaintained.
> I'd recommend learning and editing MR2 application to do what you wish
> to do, as that is the future the project has chosen.
> If however you strongly wish to edit code of the past, which we
> continue to carry for the 1.x maintenance branch alone, check out the
> branch-1 code and not trunk. Some help on it at
> On Tue, Dec 18, 2012 at 4:20 AM, Tiago Kepe <[EMAIL PROTECTED]> wrote:
> > Hi,
> > I have tried build the JobTracker.java, I cloned the hadoop-common
> > using git-clone, but I am not able to build JobTracker.
> > This class is on the directory
> > hadoop-common/hadoop-mapreduce-project/src/java/org/apache/hadoop, I
> > compile the project using Maven with command "mvn compile", but when
> > I import the mapreduce-project on Eclipse, the JobTracker.java don't
> > like source code.
> > I want to customize the JobTracker.java, but I am not getting the
> > source code on Eclipse.
> > Thank in advance.
> > --
> > Atenciosamente,
> > *Tiago Rodrigo Kepe*
> > Graduando em Ciência da Computação - UFPR Analista de
> > Desenvolvimento na Siemens Enterprise.
> Harsh J
*Tiago Rodrigo Kepe*
Graduando em Ciência da Computação - UFPR Analista de Desenvolvimento na