Granular flow simulations randomly exit on signal 1 (Hangup)

Hi all,

I am running simulations of granular flows under plane shear. When my simulations exceed about 10^7 timesteps, they shut down randomly with the following error message:

mpirun noticed that process rank 7 with PID 3389 on node master exited on signal 1 (Hangup).

There is no error message printed in the log file, and the thermo values seem to behave normally. In addition, the crashes happens very randomly: sometimes the simulation is able to complete 10^9 timesteps, sometimes it shuts down around 10^7 or 10^8 timestemps. The problem happens both in serial and parallel (the latter using anywhere from 4 to 9 cores).

The input file seems physically well-posed (since the simulations sometimes complete just fine), so I am at a loss regarding the origin of the issue and would appreciate any help. I have uploaded here the lammps input file, the restart file as well as the console output from one of the failed runs.

Thank you,

Is this on a cluster or on your own desktop? Perhaps your runs exceed the wall time limit.

Thanks for your reply! This is on a cluster, but which does not have a wall time limit.