Daniel:
Can you post your script to the list?
No problem. Here it is.
One possibility: you’re using a very small lattice spacing; things might be blowing up on you because you have too many atoms too close together. . .
I also suspected that as the problem.
v/r,
in.test_nersc_minimize (1.49 KB)
As AEI pointed out, you lattice spacing is very small, which produces a very large number of neighbors per atom. You compounded that by creating a relatively large number of atoms. When I drop the region dimensions in your script to 10x10x10, LAMMPS figures out the problem before a system error occurs:
LAMMPS (7 Jul 2009)
Lattice spacing in x,y,z = 0.1 0.1 0.1
Created orthogonal box = (0 0 0) to (1 1 1)
1 by 1 by 1 processor grid
Created 1000 atoms
WARNING: Resetting reneighboring criteria during minimization
Setting up minimization …
ERROR on proc 0: Neighbor list overflow, boost neigh_modify one or page
Aidan