An unforeseen killing while running

ASPECT
1

test.prm (9.2 KB)
After running for about 10 hours, the line was killed using 2.5.0 (not debug) with text:

----------------------------------------------------
Exception on MPI process <51> while running postprocessor <N6aspect11Postprocess18HeatFluxStatisticsILi2EEE>: 

--------------------------------------------------------
An error occurred in line <624> of file </home/fuwuqi/aspect/base/candi/tmp/unpack/deal.II-v9.6.1/source/fe/mapping_q.cc> in function
    dealii::Point<dim> dealii::MappingQ<dim, spacedim>::transform_real_to_unit_cell(const typename dealii::Triangulation<dim, spacedim>::cell_iterator&, const dealii::Point<spacedim>&) const [with int dim = 2; int spacedim = 2; typename dealii::Triangulation<dim, spacedim>::cell_iterator = dealii::TriaIterator<dealii::CellAccessor<2, 2> >]
The violated condition was: 
    numbers::is_finite(p_unit[0])
Additional information: 
    Computing the mapping between a real space point and a point in
    reference space failed, typically because the given point lies outside
    the cell where the inverse mapping is not unique.

I can do nothing about it and I’m longing for help.

2

@KIVY Does the error message have any additional information, say a backtrace where that function was called?
Best
W.

3

Dear sir, the error message is listed by following:

An error occurred in line <624> of file </home/fuwuqi/aspect/base/candi/tmp/unpack/deal.II-v9.6.1/source/fe/mapping_q.cc> in function
    dealii::Point<dim> dealii::MappingQ<dim, spacedim>::transform_real_to_unit_cell(const typename dealii::Triangulation<dim, spacedim>::cell_iterator&, const dealii::Point<spacedim>&) const [with int dim = 2; int spacedim = 2; typename dealii::Triangulation<dim, spacedim>::cell_iterator = dealii::TriaIterator<dealii::CellAccessor<2, 2> >]
The violated condition was: 
    numbers::is_finite(p_unit[0])
Additional information: 
    Computing the mapping between a real space point and a point in
    reference space failed, typically because the given point lies outside
    the cell where the inverse mapping is not unique.

Stacktrace:
-----------
#0  /home/fuwuqi/aspect/base/candi/deal.II-v9.6.1/lib/libdeal_II.so.9.6.1: dealii::MappingQ<2, 2>::transform_real_to_unit_cell(dealii::TriaIterator<dealii::CellAccessor<2, 2> > const&, dealii::Point<2, double> const&) const
#1  ./aspect: aspect::MaterialModel::Rheology::Elasticity<2>::fill_reaction_outputs(aspect::MaterialModel::MaterialModelInputs<2> const&, std::vector<double, std::allocator<double> > const&, aspect::MaterialModel::MaterialModelOutputs<2>&) const
#2  ./aspect: aspect::MaterialModel::ViscoPlastic<2>::evaluate(aspect::MaterialModel::MaterialModelInputs<2> const&, aspect::MaterialModel::MaterialModelOutputs<2>&) const
#3  ./aspect: std::vector<std::vector<std::pair<double, double>, std::allocator<std::pair<double, double> > >, std::allocator<std::vector<std::pair<double, double>, std::allocator<std::pair<double, double> > > > > aspect::Postprocess::internal::compute_heat_flux_through_boundary_faces<2>(aspect::SimulatorAccess<2> const&)
#4  ./aspect: aspect::Postprocess::HeatFluxStatistics<2>::execute[abi:cxx11](dealii::TableHandler&)
#5  ./aspect: aspect::Postprocess::Manager<2>::execute[abi:cxx11](dealii::TableHandler&)
#6  ./aspect: aspect::Simulator<2>::postprocess()
#7  ./aspect: aspect::Simulator<2>::run()
#8  ./aspect: void run_simulator<2>(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool, bool, bool)
#9  ./aspect: main
--------------------------------------------------------

Aborting!
----------------------------------------------------
--------------------------------------------------------------------------
MPI_ABORT was invoked on rank 51 in communicator MPI_COMM_WORLD
with errorcode 1.

NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.
4

Dear sir, the message was listed above. I’m sorry to miss “at”.

5

@KIVY I don’t know for sure, but typically this sort of error happens because you have a very distorted mesh. I assume you are using a free surface and end up with a mesh that has poorly shaped cells? Perhaps you can show us the mesh you have in the last time step before the program crashes?

If I am right with my hypothesis, the standard remedy is to use surface diffusion to smooth out the mesh.

Best
W.

6

u1
u1700×515 33.3 KB

u2
u2881×644 103 KB

Like these pictures…
And thank you for your advice, I’ll try it and if it works, you help me a lot.