Re: [AMBER] cpptraj.MPI error

From: Daniel Roe <daniel.r.roe.gmail.com>
Date: Mon, 7 Oct 2019 15:42:46 -0400

The problem is that cpptraj.MPI cannot find the same MPI it was compiled with.

Based on your script, try adding:

export LD_LIBRARY_PATH=$MPI_HOME/lib:$LD_LIBRARY_PATH

below the export MPI_HOME

-Dan

On Mon, Oct 7, 2019 at 3:34 PM Debarati DasGupta
<debarati_dasgupta.hotmail.com> wrote:
>
> I have a job submission script for analysis of my trajectory files.
> I am using cpptraj.MPI for the analysis, but when I submit my script, I get an error as “Working Directory = / ACN_H2O_5-95vv_layered/cumulative_prob_1P3/RUN1
> Hostname = grape6
> Date = Mon Oct 7 15:25:27 EDT 2019
> cpptraj.MPI: error while loading shared libraries: libmpicxx.so.12: cannot open shared object file: No such file or directory
> cpptraj.MPI: error while loading shared libraries: libmpicxx.so.12: cannot open shared object file: No such file or directory”
>
> I tried adding source Amber.sh in my bashrc but it did not help!
>
> My script contents are as follows:
> #! /bin/bash
> #SBATCH -J updated_1P3_jobs
> #SBATCH --partition=cpu_partition
> #SBATCH -n 4 # This means 4 processors
>
> ###############################################################
> export submit_dir=/ACN_H2O_5-95vv_layered/cumulative_prob_1P3/$1
> cd $submit_dir
> echo Working Directory = $submit_dir
> echo Hostname = `hostname -s`
> echo Date = `date`
>
>
> export AMBERHOME=/usr/local/amber18
> export CUDA_HOME=/usr/local/cuda-9.2
> export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:$CUDA_HOME/lib64
> export MPI_HOME=/usr/local/mpich-3.2.1
>
> while read x ;
> do
>
> $MPI_HOME/bin/mpiexec -n 4 cpptraj.MPI -i "$x"
> done<$2
> ~
> ~
> ~
> _______________________________________________
> 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 Mon Oct 07 2019 - 13:00:03 PDT
Custom Search