Problems with simulation crashs

Submitted by mkoester on Mon, 05/31/2010 - 17:56

Hi Everybody,

I'm running a larger big simulation (~150k particles) with granular interactions. Most other simulations wotk fine on my system, but this one simulations crash allways after exactly 320000 steps with this errors:

[prometheus:28676] *** Process received signal ***
[prometheus:28676] Signal: Bus error (10)
[prometheus:28676] Signal code: (2)
[prometheus:28676] Failing at address: 0xc
[prometheus:28674] [ 0] 2 libSystem.B.dylib 0x94fe42bb _sigtramp + 43
[prometheus:28674] [ 1] 3 ??? 0xffffffff 0x0 + 4294967295
[prometheus:28674] [ 2] 4 libSystem.B.dylib 0x94fade67 vfprintf_l + 100
[prometheus:28674] [ 3] 5 libSystem.B.dylib 0x94faddf9 fprintf + 75
[prometheus:28674] [ 4] 6 lmpc 0x000df969 _ZN9LAMMPS_NS15FixTriNeighlist9pre_forceEi + 489
[prometheus:28674] [ 5] 7 lmpc 0x0010d1b1 _ZN9LAMMPS_NS6Modify9pre_forceEi + 55
[prometheus:28674] [ 6] 8 lmpc 0x001bbdfa _ZN9LAMMPS_NS6Verlet3runEi + 540
[prometheus:28674] [ 7] 9 lmpc 0x001a22ad _ZN9LAMMPS_NS3Run7commandEiPPc + 2105
[prometheus:28674] [ 8] 10 lmpc 0x000fcc58 _ZN9LAMMPS_NS5Input15execute_commandEv + 5692
[prometheus:28674] [ 9] 11 lmpc 0x000fd992 _ZN9LAMMPS_NS5Input4fileEv + 758
[prometheus:28676] [ 0] 2 libSystem.B.dylib 0x94fe42bb _sigtramp + 43
[prometheus:28676] [ 1] 3 ??? 0xffffffff 0x0 + 4294967295
[prometheus:28676] [ 2] 4 libSystem.B.dylib 0x94fade67 vfprintf_l + 100
[prometheus:28676] [ 3] 5 libSystem.B.dylib 0x94faddf9 fprintf + 75
[prometheus:28676] [ 4] 6 lmpc 0x000df969 _ZN9LAMMPS_NS15FixTriNeighlist9pre_forceEi + 489
[prometheus:28676] [ 5] 7 lmpc 0x0010d1b1 _ZN9LAMMPS_NS6Modify9pre_forceEi + 55
[prometheus:28676] [ 6] 8 lmpc 0x001bbdfa _ZN9LAMMPS_NS6Verlet3runEi + 540
[prometheus:28676] [ 7] 9 lmpc 0x001a22ad _ZN9LAMMPS_NS3Run7commandEiPPc + 2105
[prometheus:28676] [ 8] 10 lmpc 0x000fcc58 _ZN9LAMMPS_NS5Input15execute_commandEv + 5692
[prometheus:28676] [ 9] 11 lmpc 0x000fd992 _ZN9LAMMPS_NS5Input4fileEv + 758
[prometheus:28676] [10] 12 lmpc 0x001026c0 main + 96
[prometheus:28676] [11] 13 lmpc 0x000020a6 start + 54
[prometheus:28676] *** End of error message ***

Has someone ever noticed this behaviour? Do you think its a liggghts problem or something more related to my system?

regards,
martin

ckloss's picture

ckloss | Mon, 05/31/2010 - 18:13

Hi martin,

seems like some debugging code was left unfortunately. That will be fixed in the next release.

In the meantime, go to fix_tri_neighlist.cpp and change the following line (around line 36):
#define DEBUGMODE_TRINEIGH false

Best regards,
Christoph