>>
>>> Can¹t seem to get past this heap size error, any ideas where to look? Below
>>> are my heap size settings, at least the ones I attempted to increase.
>>>
>>> Thanks in advance for any thoughts.
>>>
>>>
>>> # The maximum amount of heap to use, in MB. Default is 1000.
>>> export HADOOP_
Ronnie,
It looks like you found out, the JobClient is only useful
for connecting within the scope of the jobs you submit through it. If
you use JobClient.getAllJobs(); it will only return jobs within the
scope of that instance and not jobs that were sent by another
application or through hadoo
Moving it to mapreduce-user.
Ronnie,
Is jobclient.getalljobs() something that you are looking for?
thanks
mahadev
On Sat, Nov 12, 2011 at 7:08 AM, Ronnie Dove wrote:
> Hello,
>
> Is it possible to connect to an already running JobTracker via socket through
> the Hadoop Java API? I see that
Arun,
Take a look at the default scheduler of the JobTracker to get started:
JobQueueTaskScheduler.
On 12-Nov-2011, at 12:36 PM, arun k wrote:
> Hi !
>
> I want to understand how tasks are assigned to a taskracker/node in mumak
> then change it .
>
> For that i see that in SimulatorJobTrac
