Segmentation fault

Dear LAMMPS users,

I have got a problem of segmentation fault runing equilibration of water films on gold surfaces at 400K. I performed the cases with 96 cores on Stampede from TACC. The error message is shown below

“[c514-003.stampede.tacc.utexas.edu:mpi_rank_15][error_sighandler] Caught error: Segmentation fault (signal 11)”

However, I would not get this error if I equilibrated the system at 300K. I wonder whether you have any clue about how to deal with the segmentation fault.

Bests,

Han Hu

Complex Fluids and Multiphase Transport Lab
Mechanical Engineering and Mechanics Department
Drexel University
Hess Engineering Research Laboratories
[email protected]…1685…, [email protected]…24…

Dear LAMMPS users,

I have got a problem of segmentation fault runing equilibration of water
films on gold surfaces at 400K. I performed the cases with 96 cores on
Stampede from TACC. The error message is shown below

"[c514-003.stampede.tacc.utexas.edu:mpi_rank_15][error_sighandler] Caught
error: Segmentation fault (signal 11)"

However, I would not get this error if I equilibrated the system at 300K. I
wonder whether you have any clue about how to deal with the segmentation
fault.

segmentation fault is probably the most nondescript error that a
program an cause. it is a symptom of some underlying error, which can
have many reasons ranging from bad dynamics or bad input parameters to
a bug in the code.

if you have to similar inputs out of which one triggers a segfault and
the other doesn't, then you should first try to simplify each of them
(i.e. remove all options/features that have no impact on the segfault)
and then with these trimmed down inputs start from one and gradually
transition to the other until you see which change exactly causes the
segfault and then come back to this list posting the resulting input
with the offending line/lines highlighted and people can comment on
where the root of this lies.

axel.

Hi Axel,

Thanks for your suggestion. I will try it right now.