[gmx-users] errors depending on (lack of) solvent

gianluca santarossa gianluca.santarossa at unimib.it
Fri Jan 23 09:16:01 CET 2004

Dear Users,

I'm trying to run the gromacs tutorial demo
using lammpi on more than one PC.
Everything works fine if I add water molecules to the system using
But if I try to run a simulation in vacuo (simply removing the '-cs'
option from genbox) mpirun exits prematurely:

MPI process rank 1 (n0, p27449) caught a SIGSEGV.
MPI process rank 2 (n1, p25070) caught a SIGSEGV.
MPI process rank 5 (n2, p14853) caught a SIGSEGV.
MPI process rank 6 (n3, p29311) caught a SIGSEGV.

This is my script (Every step preciding mdrun_mpi works fine):


pdb2gmx -f ${MOL}.pdb -o ${MOL}.gro -p ${MOL}.top

editconf -f ${MOL}.gro -o ${MOL}.gro -d 1.5
#genbox -cp ${MOL}.gro  -cs -o ${MOL}_new.gro -p ${MOL}.top     # This
one works fine

genbox -cp ${MOL}.gro -o ${MOL}_new.gro -p
${MOL}.top                                 # This one doesn't work. It
lacks '-cs' option

grompp -np 4 -f em -c ${MOL}_new -p $MOL -o ${MOL}

mpirun -np 4 /usr/local/bin/mdrun_mpi -nice 4 -s ${MOL} -o ${MOL} -c
${MOL}_last -v

By now I need to run simulations in vacuo, so I cannot use the -cs
option in genbox.
Does anyone know the cause of these errors?


Gianluca Santarossa
Molecular Modelling Laboratory
Dept. of Biothecnology and Biosciences
Universty of Milano Bicocca
P.zza della Scienza 2
20126 Milan, Italy
Tel: +39-02-6448-3475
gianluca.santarossa at unimib.it

More information about the gromacs.org_gmx-users mailing list