[aspect-devel] Trilinos problems ( I guess )

[Timo Heister]

Jonathan,

I haven't seen something like that before. The only thing I can think
of is testing what library is breaking things: start by going back in
time with ASPECT, then deal.II. Annoying. :(

On Thu, Sep 10, 2015 at 7:37 AM, Jonathan Perry-Houts
<jperryh2 at uoregon.edu> wrote:
> I just updated / recompiled deal.ii and aspect on my cluster and now
> aspect is failing every time with a convergence error.
>
> I attached all the outputs from running convection-box.prm to this email
> (hopefully that works with the listserv). The important part is this
> stuff from stderr:
>
> ----------- 8< ---- CUT HERE ------ 8< -----
> The iterative Stokes solver did not converge. It reported the following
> error:
> ....
> The iterative (top left) solver in BlockSchurPreconditioner::vmult did
> not converge. It reported the following error:
> ....
> AztecOO::Iterate error code -2: numerical breakdown
> ----------- 8< ---- CUT HERE ------ 8< -----
>
> This happens for all models I've tried.
>
> I've never run into this on my desktop, or on the cluster before. I'm
> not sure what could have changed to affect the Stokes solver, but I'm
> not doing anything different in my own workflow so it must have been
> some system library that Trilinos uses, I guess... Any ideas which one
> it could be?
>
> For reference, I'm using Trilinos 11.12.1, deal.ii development source,
> aspect development source.
>
> Thanks in advance!
>
> Cheers,
> Jonathan
>
> --
> Jonathan Perry-Houts
> Ph.D. Candidate
> Department of Geological Sciences
> 1272 University of Oregon
> Eugene, OR 97403-1272
>
>
>
>
> _______________________________________________
> Aspect-devel mailing list
> Aspect-devel at geodynamics.org
> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/aspect-devel