Is there anyway to query the JobTracker remotely for the current
FairScheduler pools and configuration without having to access the
pools.xml or screen scraping the FairScheduler web page? pools.xml wont
contain dynamically created pools. I am using hadoop 1.0.3.

Thanks

Unfortunately there's no easy way to do that. It doesn't seem to have
any metrics exposed in any other form aside of the web servlet it
offers.

However, the FairScheduler does have an eventlog configuration
parameter you can use. It logs some events onto a file, which you can
perhaps parse? Doesn't give everything though.

On Thu, Aug 23, 2012 at 8:43 AM, Jim Donofrio <[EMAIL PROTECTED]> wrote:
> Is there anyway to query the JobTracker remotely for the current
> FairScheduler pools and configuration without having to access the pools.xml
> or screen scraping the FairScheduler web page? pools.xml wont contain
> dynamically created pools. I am using hadoop 1.0.3.
>
> Thanks

--
Harsh J