Thanks a lot,
I tried several tests, and on this one:
#!/bin/sh
#
#PBS -N test_equil
#PBS -q all_l_p
#PBS -l select=3:ncpus=12:mpiprocs=12
PBS_O_WORKDIR=$HOME/projects/HayDvir/Y847C/test
cd $PBS_O_WORKDIR
mpirun -hostfile $PBS_NODEFILE pmemd.MPI -O -i equil.in -p 3SO6_Y847C_clean.prmtop -c 3SO6_Y847C_clean_heat.rst
The job runs producing the default ouptut files, BUT the ns/day is exactly the same as in 1 node, so I guess output files are writting one on top of the others?
Although the qstat command produces the following ouptut:
Req'd Req'd Elap
Job ID Username Queue Jobname SessID NDS TSK Memory Time S Time
58522.admin fglaser all_l_p test_equil 20115 3 36 23gb 24:00 R 01:07
Which I understand means it is using 36 cpus???
Confusing...
Thanks a lot,
Fabian
_______________________________
Fabian Glaser, PhD
Bioinformatics Knowledge Unit,
The Lorry I. Lokey Interdisciplinary
Center for Life Sciences and Engineering
Technion - Israel Institute of Technology
Haifa 32000, ISRAEL
fglaser.technion.ac.il
Tel: +972 4 8293701
Fax: +972 4 8225153
On Dec 18, 2012, at 4:18 PM, David A Case wrote:
> On Tue, Dec 18, 2012, Fabian Glaser wrote:
>>
>> HYDU_create_process (./utils/launch/launch.c:94): execvp error on file
>> 3SO6_clean_prod_3.rst (No such file or directory)
>
> I don't have the answer, but the error is quite odd: Amber would never be
> doing an execvp on the rst file. Have you really triple-checked that the
> *identical* script works on one node but fails on two? For example, if your
> script had something like this:
>
> mpirun -np 8 /blah/blah/sander.MPI -O -i mdin -p prmtop -r \
> 3SO6_clean_prod_3.rst -c incprd ....
>
> And the backslash at the end of the first line got lost, the script would end
> up trying to execute a file called "3SO6_clean_prod_3.rst".
>
> Of course, this is a made-up example. But my point is that you may be looking
> in the wrong place for the error. Try moving arguments around. Try a short
> run with no named restart file. Try a run where one of the input files is
> deliberately mis-typed. Try a sander.MPI run with no arguments whatsoever.
>
> The idea is to narrow down the error, which may have nothing to do with not
> finding the output file on some node. Can you get *any* sander.MPI job to run
> on more than one node?
>
> ...good luck...dac
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Dec 19 2012 - 05:00:03 PST
