[CIG-SHORT] PETSc error

[Brad Aagaard]

Jiangning-

The problem could be in the mesh itself, the fault surface, or the parameters. 
I would start by running the problem without a fault specified and see if you 
get the same error message. Make sure the material 'id' values for the bulk 
materials and faults are all unique and match those in the mesh generation 
(see the PyLith manual for details).

Brad


On Tuesday 04 December 2007, Matthew Knepley wrote:
> It appears this is a problem with the fault definition. There should
> be a diagonal
> element (even if it is 0) if you want to use ILU, or almost anything
> else. Brad, doesn't
> the Lagrange stuff set a 0 diagonal?
>
>   Thanks,
>
>      Matt
>
> On Dec 4, 2007 12:22 PM, Jiangning Lu <jlu at chandler.mit.edu> wrote:
> > SORRY, wrong file.
> >
> > Here is the log file.
> >
> > Thanks!
> > Jiangning
> >
> > Quoting Matthew Knepley <knepley at gmail.com>:
> > > I don't see any errors in run.log?
> > >
> > >  Matt
> > >
> > > On Dec 3, 2007 11:27 PM, Jiangning Lu <jlu at chandler.mit.edu> wrote:
> > >> Hi all,
> > >>
> > >> Finally I installed PyLith successfully, many thanks for all your
> > >> help!!
> > >>
> > >> I've been able to run those examples smoothly.  Now I tried to run one
> > >> of my test model. This model is based on the example in
> > >> 3d/tet4/dislocation.cfg, the
> > >> major different is that I use another mesh with complicated fault
> > >> geometry. The material is uniformly elastic and I put uniform
> > >> dislocation on the fault nodes.
> > >>
> > >> The test run did go the the very end, but a bunch of "PETSc error"s
> > >> occurred during the way.
> > >>
> > >> Attached is the log file with error info.
> > >>
> > >> Also input files for the test model are attached.
> > >> Input mesh comes from LaGriT.
> > >>
> > >> Thanks!!
> > >>
> > >> Jiangning
> > >>
> > >>
> > >> ---
> > >> Jiangning Lü
> > >> 54-521     x3-6389
> > >> Geophysics
> > >> EAPS, MIT
> > >> johnnylu at mit.edu (????
> > >> _______________________________________________
> > >> CIG-SHORT mailing list
> > >> CIG-SHORT at geodynamics.org
> > >> http://geodynamics.org/cgi-bin/mailman/listinfo/cig-short
> > >
> > > --
> > > What most experimenters take for granted before they begin their
> > > experiments is infinitely more interesting than any results to which
> > > their experiments lead.
> > > -- Norbert Wiener
> >
> > ---
> > Jiangning Lü
> > 54-521     x3-6389
> > Geophysics
> > EAPS, MIT
> > johnnylu at mit.edu (????