[Gmsh] Gmsh aborted in the middle of 3D mesh

Christophe Geuzaine geuzaine at acm.caltech.edu
Wed Oct 6 09:10:55 CEST 2004


C. D. Wang wrote:
> I tried to 3D mesh this simple cube when Gmsh was aborted suddenly. the
> .gmsh-errors file said :
> 
> Info    : Mesh 2D complete (0.055991 s)
> Info    : Mesh 3D...
> Info    : Meshing volume 1000
> Info    : CalcLocalH: 686 Points 0 Elements 1368 Surface Elements
> Info    : Check subdomain 1 / 1
> Info    : 1368 open elements
> Fatal   : ERROR: Edge 1 - 9 multiple times in surface mesh
> Info    : Writing log file '/home/cynthiaw/.gmsh-errors'
> 
> Can anyone please tell me what I did wrong? Thank you!
> 

Cynthia - Your example works just fine with Gmsh's original isotropic 3D
algorithm. With the Netgen algorithm, you need to orient all your 
surfaces with exterior pointing normals. Just replace

Surface Loop(200)={100,103,101,105,-102,-104};

with

Surface Loop(200)={-100,-103,-101,-105,102,104};

and everything should be okay.

Cheers,

Christophe

-- 
Christophe Geuzaine
Applied and Computational Mathematics, Caltech
geuzaine at acm.caltech.edu - http://geuz.org