Sim. working on 1 CPU, crashing on 2 and more CPU

LAMMPS LAMMPS Mailing List Mirror
1

Hi,

If i run my simulation on 1 CPU, its running fine. When i start simulation on more CPU i got this message:

ERROR on proc 0: Improper atoms 55 54 57 78 missing on proc 0 at step 0 (…/neigh_bond.cpp:831)

Weird part is that i got this message before first step. I even remove time integration and set “run 1” and its still crashing. On one CPU, its going for 100 000 steps and more.
Any suggestion, where is problem?​

Thank you!

Upozornění: Není-li v této zprávě výslovně uvedeno jinak, není tato e-mailová zpráva návrhem na uzavření smlouvy ani přijetím případného návrhu na uzavření smlouvy a nezakládá předsmluvní odpovědnost Ústavu chemických procesů AV ČR, v. v. i. Disclaimer: If not expressly stated otherwise, this e-mail message cannot be considered as a proposal to conclude a contract, neither the acceptance of a proposal to conclude a contract, nor does it create any pre-contractual liability on the part of the Institute of Chemical Process Fundamentals of the CAS, v. v. i.

2

Hi,

If i run my simulation on 1 CPU, its running fine. When i start simulation
on more CPU i got this message:

ERROR on proc 0: Improper atoms 55 54 57 78 missing on proc 0 at step 0
(../neigh_bond.cpp:831)

Weird part is that i got this message before first step. I even remove time
integration and set "run 1" and its still crashing. On one CPU, its going
for 100 000 steps and more.
Any suggestion, where is problem?

difficult to say without actually seeing the entire input. there are a
number of possible reasons, e.g:
- use of shrinkwrap boundary conditions with too large "free" space,
so that the shrink on the first step is too large
- incorrect definition of the improper in question, so that it reaches
very far through space and beyond the cutoff of ghost atoms
- too short communication cutoff due to a small neighbor skin and the
largest pair cutoff being shorter than the length of the improper in
question

axel.