[lammps-users] error with bonds



Dear all,







The following error message appears when I try to simulate a single molecule with 27 atoms using one or more processors for 5000 time steps. The boundaries are periodic in all directions with fix npt. Changing the number of processors to more than one, only changes the time step at which the ERROR occurs.








Any suggestions on why this occurs and how to overcome it would be very useful







Thanks



Ganesh







LAMMPS (7 Jul 2009)
n4(0): Scanning data file …
n4(0): 6 = max bonds/atom
n4(0): 6 = max angles/atom
n4(0): Reading data file …
n4(0): orthogonal box = (1.023 -3.753 -2.128) to (12.891 8.243 4.729)
n4(0): 1 by 1 by 1 processor grid
n4(0): 27 atoms
n4(0): 24 bonds
n4(0): 13 angles
n4(0): Finding 1-2 1-3 1-4 neighbors …
n4(0): 6 = max # of 1-2 neighbors
n4(0): 12 = max # of 1-3 neighbors
n4(0): 15 = max # of 1-4 neighbors
n4(0): 18 = max # of special neighbors
n4(0): Ewald initialization …
n4(0): G vector = 0.54631
n4(0): vectors: actual 1d max = 299 6 1098
n4(0): Setting up run …
n4(0): Memory usage per processor = 5.06355 Mbytes
n4(0): Step Atoms Temp Press PotEng KinEng
n4(0): 0 27 300 -332911.58 117.62282 1.0082291
n4(0): 10 27 41159.912 149477.44 -16.566268 138.32874
n4(0): 20 27 13719.382 98001.954 67.856622 46.107602
n4(0): 30 27 22412.086 -24158.414 37.904143 75.321727
n4(0): 40 27 23803.467 -11053.918 33.502415 79.997828








n4(0): 990 27 206427.59 318009.73 254.69202 693.75436
n4(0): 1000 27 232278.05 490120.07 173.33002 780.63164
n4(0): 1010 27 270434.12 456702.44 289.14311 908.86518
n4(0): ERROR on proc 0: Bond atoms 4 11 missing on proc 0 at step 1020
n4(0): [0] [MPI Abort by user] Aborting Program!
All nodes terminated successfully.






|

ganesh,

same as the previous report. it looks like your
initial structure is far from equilibrium and
thus atoms heat up very much. try to minimize
first and make sure the parameters are correct.

the error message originates from an atom
move too fast.

cheers,
  axel.