I used the 'fix multisphere/break' feature for a simple direct shear simulation on clusters. After breakage occurs, I would soon receive error message telling me that there is problem with MPI all_reduce. The simulation went well when I used 'fix multisphere' instead. I don't have any idea what went wrong with it.
Xinyi Qian | Tue, 12/28/2021 - 08:53
Additional information
The error message is as follows
-------------------------------------------------
*** An error occurred in MPI_Allreduce
*** reported by process [47638248030209,47635482279967]
*** on communicator MPI_COMM_WORLD
*** MPI_ERR_TRUNCATE: message truncated
*** MPI_ERRORS_ARE_FATAL (processes in this communicator will now abort, and potentially your MPI job)
-------------------------------------------------
The command I used for this fix style is:
fix multisphere all multisphere/break allow_group_and_set yes trigger_threshold 100 trigger_timeStep 981000 trigger_name v_devstress
where devstress is a variable I defined
mschramm | Tue, 12/28/2021 - 19:42
simple example
Hello,
Can you supply a simple script that we can test with?
Your command seems reasonable.