[filipedireito]

Hi! (And happy new year)

New year, old question. I was running a case with interFoam in parallel with 4 processors. It was fine. Then I decided to refine the mesh and the problem started. The simulations start running but at apparently random moment they stop in one of the these status (some old threads in this forum describe similar behavior):
- segmentation fault
- floating point exception (sometimes, but not always, with extremely high residuals)
- total pc freeze (have to reboot).

I then decided to test this on the damBreak case, to confirm the problem was not somewhere in my case. Running damBreak with 4 processors and 31k or 123k cells is apparently flawless. However, at 277k cells, the freezing/floating/segmentation problem appears.

I have read several webpages (most of them in CFD Online) and tried things like changing the MPI min Buffer. But I did not manage to solve the problem...

My system:
OF 4.1, systemOpenMPI
ubuntu 16.04
4720HQ, 16 gb ram, 16 gb swap

How I run:
setFields
decomposePar
mpirun -np 4 interFoam -parallel > log &