Getting error as Floating point exception (core dumped) middle main run MD simulation

User discussions
1

GROMACS version: 2019.6
GROMACS modification: Yes/No
I got this error while running for MD simulation: Floating point exception(core dumped) gmx mdrun -deffnm md_0_10 -v -ntmpi 1 -ntomp 40

Reading file md_0_10.tpr, VERSION 2019.6 (single precision)
Non-default thread affinity set, disabling internal thread affinity
Changing nstlist from 10 to 100, rlist from 1 to 1.168

Using 1 MPI thread
Using 40 OpenMP threads

1 GPU selected for this run.
Mapping of GPU IDs to the 2 GPU tasks in the 1 rank on this node:
PP:0,PME:0
PP tasks will do (non-perturbed) short-ranged interactions on the GPU
PME tasks will do all aspects on the GPU

WARNING: This run will generate roughly 7195 Mb of data

starting mdrun ‘Protein in water’
150000000 steps, 300000.0 ps.
step 200: timed with pme grid 52 52 52, coulomb cutoff 1.000: 232.9 M-cycles
step 400: timed with pme grid 44 44 44, coulomb cutoff 1.123: 249.2 M-cycles
step 600: timed with pme grid 40 40 40, coulomb cutoff 1.235: 314.3 M-cycles
step 800: timed with pme grid 42 42 42, coulomb cutoff 1.177: 344.7 M-cycles
step 1000: timed with pme grid 44 44 44, coulomb cutoff 1.123: 330.0 M-cycles
step 1200: timed with pme grid 48 48 48, coulomb cutoff 1.030: 467.1 M-cycles
step 1400: timed with pme grid 52 52 52, coulomb cutoff 1.000: 333.4 M-cycles
optimal pme grid 52 52 52, coulomb cutoff 1.000
step 60882100, will finish Sun Jan 19 17:27:41 2025./md_mdrun.sh: line 1: 2538384 Floating point exception(core dumped) gmx mdrun -deffnm md_0_10 -v -ntmpi 1 -ntomp 40
Sat Jan 18 07:12:55 PM CST 2025
thanks for your help.

2

This is either a bug or a that the system is exploding without GROMACS being able to handle it as gracefully as it should.

The GROMACS version you are using is not supported anymore. If it’s a bug it’s hopefully already fixed in a later version. So, what you can do is to

  • upgrade to GROMACS 2024 (2025 will be released soon)
  • see if you can continue from the previous checkpoint without getting the same crash
    or
  • run the equilibrations longer and/or check your barostat settings.