Case diverges after running renumberMesh in parallel
 

Hi all,

I'm using OpenFOAM-dev on a Scientific Linux cluster. Generally I run renumberMesh in parallel, after decomposePar. However, in some cases this causes my simulation to diverge. For example, if I take the DTCHullMoving tutorial and run renumberMesh after decomposePar, it crashes fairly quickly, but if I run renumberMesh before decomposePar (like in the Allrun script provided with the case) the simulation runs without any trouble.

Can anyone explain why this is happening? Is that something related to dynamic meshing, or should I change the order of running decomposePar and renumberMesh for static meshes too?

Many thanks,
Sita

Apparently I'm not the only person having issues with this. I just came across this post from October 2018: https://www.cfd-online.com/Forums/op...ivergence.html
Unfortunately this post doesn't have any replies either