[gmx-users] Range checking error
yz30 at nyu.edu
Wed Dec 12 18:52:09 CET 2007
Is there anybody have met such problem before?
I did NTP simulation for POPE system downloaded from
It can run for even 10 ns with single precision GROMACS on many
processors or with double precision GROMACS on 1 or 2 processors, but
crash at about 100ps with double
GROMACS on more than 2 processors. The same thing didn't happen for a
I don't think it can be ascribed to the initial structure, because I
used the same initial structure and input files for tests. Which one do
you prefer for MD simulation, single or double precision?
Here is the error message:
Program mdrun_mpi_d, VERSION 3.3.2
Source code file: nsgrid.c, line: 226
Range checking error:
Explanation: During neighborsearching, we assign each particle to a grid
based on its coordinates. If your system contains collisions or parameter
errors that give particles very high velocities you might end up with some
coordinates being +-Infinity or NaN (not-a-number). Obviously, we cannot
put these on a grid, so this is usually where we detect those errors.
Make sure your system is properly energy-minimized and that the potential
energy seems reasonable before trying again.
Variable ci has value 4421. It should have been within [ 0 .. 4212 ]
I compiled GROMACS on Linux x86_64 with mvapich-0.9.9, this is how I
setenv CC icc
setenv CXX icpc
setenv F77 ifort
setenv MPICC mpicc
setenv CFLAGS "-O3 -fPIC -align -Zp8 -axW -unroll -Wno-deprecated -ip"
setenv CXXFLAGS "-O3 -fPIC -align -Zp8 -axW -unroll -Wno-deprecated -ip"
setenv FFLAGS "-O3 -fPIC -align -Zp8 -axW -unroll -Wno-deprecated -ip"
setenv CPPFLAGS "-I/home/yz30/Intel/FFTW-3.1.2-Lib/include"
setenv LDFLAGS "-L/home/yz30/Intel/FFTW-3.1.2-Lib/lib -L/usr/X11R6/lib64"
--enable-double --with-fft=fftw3 \
--enable-mpi --program-suffix=_mpi_d --disable-nice
Thanks in advance.
More information about the gromacs.org_gmx-users