[aspect-devel] Aspect hangs after several time steps

[Lev Karatun]

Hi,

although setting the reaction terms helps with disabled runtime FPE, if I
try to enable them Aspect crashes immediately (see the error message
below). I can't get anything from the debugger, and the model is already
simplified to the maximum. How can I solve it?..

[titan:30224] *** Process received signal ***
> [titan:30224] Signal: Floating point exception (8)
> [titan:30224] Signal code: Invalid floating point operation (7)
> [titan:30224] Failing at address: 0x3db3041cb0
> [titan:30224] [ 0] /lib64/libpthread.so.0[0x3db380f790]
> [titan:30224] [ 1] /lib64/libc.so.6[0x3db3041cb0]
> [titan:30224] [ 2] /lib64/libc.so.6(__printf_fp+0x97e)[0x3db304a53e]
> [titan:30224] [ 3] /lib64/libc.so.6(_IO_vfprintf+0x18d0)[0x3db30458a0]
> [titan:30224] [ 4] /lib64/libc.so.6(vsnprintf+0xa2)[0x3db306f752]
> [titan:30224] [ 5] /usr/lib64/libstdc++.so.6[0x3dbd87eb4f]
> [titan:30224] [ 6]
> /usr/lib64/libstdc++.so.6(_ZNKSt7num_putIcSt19ostreambuf_iteratorIcSt11char_traitsIcEEE15_M_insert_floatIdEES3_S3_RSt8ios_baseccT_+0xd3)[0x3dbd880f23]
> [titan:30224] [ 7]
> /usr/lib64/libstdc++.so.6(_ZNKSt7num_putIcSt19ostreambuf_iteratorIcSt11char_traitsIcEEE6do_putES3_RSt8ios_basecd+0x19)[0x3dbd881249]
> [titan:30224] [ 8]
> /usr/lib64/libstdc++.so.6(_ZNSo9_M_insertIdEERSoT_+0x9f)[0x3dbd89487f]
> [titan:30224] [ 9]
> /home/lev/aspect/dealii_debug_new/lib/libdeal_II.g.so.8.3.0(_ZNK6dealii8Patterns6Double11descriptionEv+0xb6)[0x7f8886369468]
> [titan:30224] [10]
> /home/lev/aspect/dealii_debug_new/lib/libdeal_II.g.so.8.3.0(_ZN6dealii16ParameterHandler13declare_entryERKSsS2_RKNS_8Patterns11PatternBaseES2_+0x6a1)[0x7f8886370569]
> [titan:30224] [11]
> ../aspect(_ZN6aspect10ParametersILi3EE18declare_parametersERN6dealii16ParameterHandlerE+0xc4a)[0x10b4a94]
> [titan:30224] [12]
> ../aspect(_ZN6aspect9SimulatorILi3EE18declare_parametersERN6dealii16ParameterHandlerE+0x18)[0x10c35c8]
> [titan:30224] [13] ../aspect(main+0x721)[0x135ae5d]
> [titan:30224] [14] /lib64/libc.so.6(__libc_start_main+0xfd)[0x3db301ed5d]
> [titan:30224] [15] ../aspect[0xf01139]
> [titan:30224] *** End of error message ***
> Floating point exception (core dumped)


Best regards,
Lev Karatun.

2016-02-23 21:40 GMT-05:00 Timo Heister <heister at clemson.edu>:

> > Did you activate the compositional fields Timo?
>
> No, I just ran the .prm he sent.
>
> On Tue, Feb 23, 2016 at 6:18 PM, Wolfgang Bangerth <bangerth at tamu.edu>
> wrote:
> > On 02/23/2016 04:58 PM, Rene Gassmoeller wrote:
> >>
> >>
> >> The problem is that the 'Simpler' material model is not using
> >> compositional
> >> fields, and does not consider that the model might still be solved with
> >> passive compositional fields. Therefore the 'reaction_terms' variable,
> >> which
> >> describes the change of the compositional fields over one timestep
> remains
> >> unfilled and contains the uninitialized value 'NaN'.
> >
> >
> > Ah, wow, good detective work. I think that the material models should
> > actually set these reaction terms to zero, since that's the material
> > behavior they describe. This is likely an issue for not just the Simpler
> > material model.
> >
> > _______________________________________________
> > Aspect-devel mailing list
> > Aspect-devel at geodynamics.org
> > http://lists.geodynamics.org/cgi-bin/mailman/listinfo/aspect-devel
>
>
>
> --
> Timo Heister
> http://www.math.clemson.edu/~heister/
> _______________________________________________
> Aspect-devel mailing list
> Aspect-devel at geodynamics.org
> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/aspect-devel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.geodynamics.org/pipermail/aspect-devel/attachments/20160226/d602caa3/attachment.html>