[torqueusers] Re: Unexpected qsub error

[David McGiven]

I forgot to say that now (without any apparent reason) it allows me to
send jobs with 2 cpu's but not with 4. What's going on ?

David

On Thu, 16 Nov 2006 12:52:07 +0100
David McGiven <david.mcgiven at fusemail.com> wrote:

> 
> Dear Torque users,
> 
> I had to powerdown my two servers for electrical maiteinance. When I
> turned them on again, I found the following error when trying to
> submit
> jobs.
> 
> "qsub: Job exceeds queue resource limits"
> 
> But i'm asking for two processors where the machine has more than 10
> processors free.
> 
> This never happened before, and I'm sure I didn't change anything from
> the config files. Could someone please give me advice ?
> 
> The only thing I did was to qdel with "-p" option some jobs before
> shutting the server down. Can this have affected ?
> 
> Thanks in advance.
> 
> David
> 
> The script I'm trying to submit :
> 
> ---------CUT---------
> #!/bin/tcsh
> #PBS -S /bin/tcsh
> #PBS -m n
> #PBS -j oe
> #PBS -k oe
> #PBS -l nodes=2
> #PBS -M david
> #PBS -q incal
> #PBS -r n
> # name
> #PBS -N gli_g03-2
> if ($?LD_LIBRARY_PATH) setenv LD_LIBRARY_PATH /opt/soft/gaussian03
> setenv g03root "/opt/soft/gaussian03"
> setenv GAUSS_SCRDIR "/scratch"
> source $g03root/g03/bsd/g03.login
> setenv GAUSS_LFLAGS "-nodefile $PBS_NODEFILE"
> 
> g03 glico01.com > glico01.log
> ---------CUT---------
> 
> The machine where pbs_server runs is :
> 
> - Dell Poweredge, Intel Xeon 3.4 Ghz, Debian, 2.6 i686 kernel
> - Troque version : torque-2.0.0p8
> - Maui version : maui-3.2.6p14
> 
> The machine where pbs_mom runs is :
> 
> - SGI Altix 330, Intel Itanium2 1.5 Ghz SLES9 (10 CPU's)
> - Torque version : 2.0.0p7
> 
> The logfile from the pbs_server :
> 
> 11/13/2006 15:39:35;0100;PBS_Server;Req;;Type AuthenticateUser request
> received from david at machine, sock=12 
> 
> 11/13/2006 15:39:35;0100;PBS_Server;Req;;Type QueueJob request
> received
> from david at machine, sock=11 
> 
> 11/13/2006 15:39:35;0080;PBS_Server;Req;req_reject;Reject reply
> code=15036(Job exceeds queue resource limits), aux=0, type=QueueJob,
> from david at machine
> 
> The logfile from the pbs_mom :
> 
> - Nothing new is shown on the log when qsub issued.
>