Segmentation falult (core dumped

GROMACS version:
GROMACS modification: Yes/No
Here post your question
hello i am using gromacs 24 and calculating diffusion of a drug molecule byut i encountered this segmentation fault error during energy minimization can anyone help me?
gmx mdrun -nt 8 -v -nb gpu -gpu_id 0 -s em.tpr -deffnm em
:-) GROMACS - gmx mdrun, 2024.1 (-:

Executable: /usr/local/gromacs/bin/gmx
Data prefix: /usr/local/gromacs
Working dir: /home/administrator/Desktop/
Command line:
gmx mdrun -nt 8 -v -nb gpu -s em.tpr -deffnm em

Back Off! I just backed up em.log to ./#em.log.4#
Reading file em.tpr, VERSION 2024.1 (single precision)
1 GPU selected for this run.
Mapping of GPU IDs to the 1 GPU task in the 1 rank on this node:
PP:0
PP tasks will do (non-perturbed) short-ranged interactions on the GPU
PP task will update and constrain coordinates on the CPU
Using 1 MPI thread
Using 8 OpenMP threads

NOTE: The number of threads is not equal to the number of (logical) cpus
and the -pin option is set to auto: will not pin threads to cpus.
This can lead to significant performance degradation.
Consider using -pin on (and -pinoffset in case you run multiple jobs).

Back Off! I just backed up em.trr to ./#em.trr.4#

Back Off! I just backed up em.edr to ./#em.edr.4#

Steepest Descents:
Tolerance (Fmax) = 1.50000e+02
Number of steps = 5000000
Step= 0, Dmax= 1.0e-03 nm, Epot= 1.98031e+33 Fmax= 8.15762e+06, atom= 14
Segmentation fault (core dumped)

There are probably atoms overlapping causing very high forces and the extremely high potential energy in the system.

thank you for your suggestion i added h-bond constraints in mdp file and it worked perfectly.

hi MagnusL, i am trying to minimize a small molecule with 73 atom, during enrgy minimization after certain point, 1 hydrogen atom is getting attached to additional oxgyen atom present nearby. i have check the positions of these two atoms in em.gro file, its very close. is it overlapping of atoms?

Hi,
When you say that “1 hydrogen atom is getting attached to additional oxygen atom”, I presume you mean that if you visualize the molecule in a software with automatic bond detection there is a bond shown between them, right? If so, that does not mean that it is getting attached, or that they are overlapping. But it is still worrying that energy minimization would put the atoms that close to each other. That indicates a very strained, or unfavorable initial geometry.

It would be improbable that energy minimization would make two atoms actually overlap. The energy barrier for that would be very high and in that case there are definitely other problems in your structure.

@MagnusL Thanks for the clarity.
yes, may be in visualization it is appearing as a bond due to closeness of two atoms.
if i ignore this visualization and proceed further then after minimizing energy to possible minimum value, nvt run is showing that “atoms are rotating more than 30 degree(segmentation fault, core dumped)”, it means my system is not stable? how to resolve this?

@jalemkul could you please help ij this?

If you get a crash in the NVT run, after a “successful” EM run, you have probably ended up in a local unfavourable configuration. Try to see what atoms are involved in the messages you get during NVT, before crashing. Have a look at those atoms, and atoms nearby, in the initial structure and try to resolve all overlaps or unfavourable bond configurations.