"Bond atom missing in image check" in parallel run of crystal structure with bond list

Dear Axel,

I’m running a simulation of Boron Nitride monolayer (with bonds, angles, dihedrals) using PBC condiction. The job could run normally with the serial version of lammps-11Aug17. However, it gave me errors when run with the parallel version of lammps-11Aug17:

ERROR on proc 3: Bond atom missing in image check (…/domain.cpp:766)
ERROR on proc 11: Bond atom missing in image check (…/domain.cpp:766)
application called MPI_Abort(MPI_COMM_WORLD, 1) - process 11

I attached my input, datafile, and output files for your reference. Could you please help me to figure out the problem?

I appreciate your help!

Thanks,
Jibao

in.adopted (748 Bytes)

readme (188 Bytes)

BN_4x4.data (17.3 KB)

job.sh (112 Bytes)

BNC.tersoff (5.48 KB)

slurm-4020882.out (2.91 KB)

Dear Axel,

I'm running a simulation of Boron Nitride monolayer (with bonds, angles,
dihedrals) using PBC condiction. The job could run normally with the serial
version of lammps-11Aug17. However, it gave me errors when run with the
parallel version of lammps-11Aug17:
ERROR on proc 3: Bond atom missing in image check (../domain.cpp:766)
ERROR on proc 11: Bond atom missing in image check (../domain.cpp:766)
application called MPI_Abort(MPI_COMM_WORLD, 1) - process 11
...

I attached my input, datafile, and output files for your reference. Could
you please help me to figure out the problem?

​the problem is, that you have no pair style defined. without a pair style,
several properties that use the associated non-bond​ cutoff as basis, e.g.
the communication cutoff, as not initialized and thus fall back to much
smaller values. are you sure that the model is correct? it is very unusual
to have a conventional force field parameterization without non-bonded
interactions.

axel.

Hi Axel,

Thanks to your help, the problem was solved.
As for the model, I knew it’s very unusual to have a conventional force field without non-bonded interactions. It was just for test. I should have added the pair interactions if I noticed that the problem was due to lack of pair style.
I appreciate your help!

thanks,
Jibao