Re: [HTCondor-users] No Starter found to Run

[Steven Timm <timm@xxxxxxxx>]

Look on the compute nodes at the StartLog, that will give
you some more clues.. If you don't see anything
set the logging of the startd and starter to D_FULLDEBUG.
I've never seen this problem happen before but it should be possible
to figure out with the right debug level.

Steve Timm


On Tue, 18 Jun 2013, Vishal Shah wrote:

> Hello,
>
> I am having an issue configuring condor to run jobs. Currently there is a
> node that acts as a head node from which the jobs are submitted and where
> all of the nodes in the compute unit can be seen; however, when submitting
> a job, the logs in the head node indicate that the job has been submitted,
> but the job does not run. The queue shows that the job has been submitted;
> however the state is perpetually pending. The following is a snippet of
> the StartLog on the compute node:
>
> 06/18/13 14:07:19 slot1: Received match
> <10.144.6.164:9532>#1371557953#351#...
> 06/18/13 14:07:19 slot1: State change: match notification protocol
> successful
> 06/18/13 14:07:19 slot1: Changing state: Unclaimed -> Matched
> 06/18/13 14:07:19 slot1: No starter found to run this job!  Is something
> wrong with your Condor installation?
> 06/18/13 14:07:19 slot1: Request to claim resource refused.
> 06/18/13 14:07:19 slot1: State change: claiming protocol failed
> 06/18/13 14:07:19 slot1: Changing state: Matched -> Owner
> 06/18/13 14:07:19 slot1: State change: IS_OWNER is false
> 06/18/13 14:07:19 slot1: Changing state: Owner -> Unclaimed
>
> Does anybody have insight into this issue?
>
> Thanks,
> Vishal

------------------------------------------------------------------
Steven C. Timm, Ph.D  (630) 840-8525
timm@xxxxxxxx  http://home.fnal.gov/~timm/
Fermilab Computing Division, Scientific Computing Facilities,
Grid Facilities Department, FermiGrid Services Group, Group Leader.
Lead of FermiCloud project.