[SOLVED] TORQUE / mpirun - error while loading shared libraries
 

UPDATE: ...


It appears that the mpirun I was using was not compiled with TORQUE support, or at least it was necessary to add the option "-x LD_LIBRARY_PATH".


Case closed.

__________________________________________________ ___________________


Hi


I have installed OpenFOAM-v1606+ on a cluster under my home directory.

It runs fine in parallel on the local node. but when I "qsub" a script containing the "mpirun" command to our cluster I get the following error

************************************************** *********
${HOME}/OpenFOAM/OpenFOAM-v1606+/platforms/linux64GccDPInt32Opt/bin/interFoam: error while loading shared libraries: libimmiscibleIncompressibleTwoPhaseMixture.so: cannot open shared object file: No such file or directory
************************************************** *********

When I qsub in interactive mode, I see that the directory containing the shared library is in my LD_LIBRARY_PATH

My submission script is given below. I could try installing the OpenFOAM foundation version, but does anyone have any idea why I am hitting the issue.

Thanks.


************************************************** **********
#!/bin/csh
# declare a name for this job to be sample_job
#PBS -N par_db
# request a total of 16 processors for this job (2 nodes and 8 processors per node)
#PBS -l nodes=2: ppn=8
# combine PBS standard output and error files
#PBS -j oe

set solver=interFoam

set fp2solver="${HOME}/OpenFOAM/OpenFOAM-v1606+/platforms/linux64GccDPInt32Opt/bin/"${solver}

# set up the OpenFOAM environment
source ${HOME}/OpenFOAM/OpenFOAM-v1606+/etc/cshrc .

#change to the directory where you submitted the job
cd $PBS_O_WORKDIR

mpirun -machinefile hosts -np 16 $fp2solver -parallel

# exit!
exit 0
************************************************** **********