LAMMPS WWW Site - LAMMPS Documentation - LAMMPS Mailing List Archives
Re: [lammps-users] ReaxFF MD simulation problem
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [lammps-users] ReaxFF MD simulation problem

From: Axel Kohlmeyer <akohlmey@...24...>
Date: Wed, 15 Nov 2017 08:07:06 -0500

On Wed, Nov 15, 2017 at 1:16 AM, higuk <higuk@...7236.....> wrote:

Dear Axel


Thanks to your reply.

Based on your advice, I have installed lammps new version (lammps-31Mar17) and I have run same script.

However, I’m faced with same problems.

Exactly speaking, the simulation is normally finished, but the calculation does not proceed.

I checked following message in my log file.

486.933u 3.011s 0:17.77 2757.1% 0+0k 8512+38960io 883pf+0w

​the attached input cannot be run in parallel (with 28 ​CPUs) as you claimed. it errors out immediately (see below), suggesting, that there is something wrong with your restrain fixes. most likely you are misunderstanding how they work, as your input seems to be hinting at that you are assuming those are based on atom types, while fix restrain works on atom ids.
it can be easily seen that atoms 1 and 5 are more than 12 \AA (the master communication cutoff) apart and thus causing the error. increasing that cutoff to make the simulation continue results in a non-physical structure.


[akohlmey@...7011... sm]$ mpirun -np 28 ~/compile/lammps/src/lmp_omp -in
LAMMPS (23 Oct 2017)
OMP_NUM_THREADS environment is not set. Defaulting to 1 thread. (../comm.cpp:90)
  using 1 OpenMP thread(s) per MPI task
Reading data file ...
  orthogonal box = (0 0 0) to (31.8535 31.8535 31.8535)
  2 by 2 by 7 MPI processor grid
  reading atoms ...
  3328 atoms
Finding 1-2 1-3 1-4 neighbors ...
  special bond factors lj:   0          0          0         
  special bond factors coul: 0          0          0         
  0 = max # of 1-2 neighbors
  0 = max # of 1-3 neighbors
  0 = max # of 1-4 neighbors
  1 = max # of special neighbors
Warning: changed valency_val to valency_boc for X
Neighbor list info ...
  update every 1 steps, delay 10 steps, check yes
  max neighbors/atom: 2000, page size: 100000
  master list distance cutoff = 12
  ghost atom cutoff = 12
  binsize = 6, bins = 6 6 6
  2 neighbor lists, perpetual/occasional/extra = 2 0 0
  (1) pair reax/c, perpetual
      attributes: half, newton off, ghost
      pair build: half/bin/newtoff/ghost
      stencil: half/ghost/bin/3d/newtoff
      bin: standard
  (2) fix qeq/reax, perpetual, copy from (1)
      attributes: half, newton off, ghost
      pair build: copy
      stencil: none
      bin: none
Setting up Verlet run ...
  Unit style    : real
  Current step  : 0
  Time step     : 1
ERROR on proc 14: Restrain atoms 1 5 missing on proc 14 at step 0 (../fix_restrain.cpp:235)
Last command: run 10000
application called MPI_Abort(MPI_COMM_WORLD, 1) - process 14



I am looking for your reply.





From: Axel Kohlmeyer [mailto:akohlmey@...24...]
Sent: Wednesday, November 15, 2017 10:24 AM
To: higuk <higuk@...7234...>
Cc: LAMMPS Users Mailing List <>
Subject: Re: [lammps-users] ReaxFF MD simulation problem




On Tue, Nov 14, 2017 at 7:55 PM, higuk <higuk@...7234...> wrote:

Dear All lammps users



I am trying to run simulation about the reaction of alkoxysilanes using ReaxFF.

I’m faced with problems during simulations.

When I’m performing ReaxFF MD using attached input script, the simulation stopped without any error messages. (There are 3328 atoms in the input structure.)

However, when I’m running the simulation with a 1/2 scale structure, this simulation proceeds normally.

I used lammps-10Aug15 version and 28 processes. (SMP cores)


How can I solve this problem?


your LAMMPS version is over 2 years old. you should upgrade. there have been many bugfixes and improvements since. nobody will look into fixing problems without confirmation that the problem hasn't been fixed already.

also, LAMMPS doesn't just stop without any kind of error message. it may just be, that you didn't look in the right place.

also, your input deck is missing the force field file your are using, so nobody can reproduce the run.





Pleases Help me.


Check out the vibrant tech community on one of the world's most
engaging tech sites,!
lammps-users mailing list



Dr. Axel Kohlmeyer  akohlmey@...24...
College of Science & Technology, Temple University, Philadelphia PA, USA
International Centre for Theoretical Physics, Trieste. Italy.

Dr. Axel Kohlmeyer  akohlmey@...24...
College of Science & Technology, Temple University, Philadelphia PA, USA
International Centre for Theoretical Physics, Trieste. Italy.