Hello Guys.
I am trying to run lammps on parallel but I am facing some problem. I am sure that the paths are correct and the executables are working because I tried the lmp_mpi excitable separately. I am running my job using slurm. My batch file looks like the he following.
#!/bin/bash -x
HOSTFILE=/tmp/hosts.$SLURM_JOB_ID
srun hostname -s > $HOSTFILE
if [ -z “$SLURM_NPROCS” ] ; then
if [ -z “$SLURM_NTASKS_PER_NODE” ] ; then
SLURM_NTASKS_PER_NODE=1
fi
SLURM_NPROCS=$(( $SLURM_JOB_NUM_NODES * $SLURM_NTASKS_PER_NODE ))
fi
/usr/local/mpich/latest/ch_p4/bin/mpirun -machinefile $HOSTFILE -np $SLURM_NPROCS /fasttmp/sengaa/400/lmp_mpi -in in.400
I am running the job using the following command batch - p normal -np 24 slurmjob.sh. I get the following error message in the output file. invalid command line argument (…/lammps.cpp 246). I want to know what is the reason for this error because when I run the same job in serial it is working. Is it something related to my cluster, MPICH or a bug within LAMMPS
Thanks
Sorry I meant I use the following command for running the job batch -p normal - n 24 slurmjob.sh
thanks
Hello Guys.
I am trying to run lammps on parallel but I am facing some problem. I am
sure that the paths are correct and the executables are working because I
tried the lmp_mpi excitable separately. I am running my job using slurm. My
batch file looks like the he following.
#!/bin/bash -x
HOSTFILE=/tmp/hosts.$SLURM_JOB_ID
srun hostname -s > $HOSTFILE
if [ -z "$SLURM_NPROCS" ] ; then
if [ -z "$SLURM_NTASKS_PER_NODE" ] ; then
SLURM_NTASKS_PER_NODE=1
fi
SLURM_NPROCS=$(( $SLURM_JOB_NUM_NODES * $SLURM_NTASKS_PER_NODE ))
fi
/usr/local/mpich/latest/ch_p4/bin/mpirun -machinefile $HOSTFILE -np
$SLURM_NPROCS /fasttmp/sengaa/400/lmp_mpi -in in.400
I am running the job using the following command batch - p normal -np 24
slurmjob.sh. I get the following error message in the output file. invalid
command line argument (../lammps.cpp 246). I want to know what is the reason
for this error because when I run the same job in serial it is working. Is
it something related to my cluster, MPICH or a bug within LAMMPS
this has to be something due to your local machine configuration.
your best chance to get help is to contact your local system managers
and work with them to get your simulations running.
axel.