Re: [Libmesh-users] problems with running libmesh in parallel

[Michael P. <mpo...@pu...>]

On 09/06/2017 10:57 AM, Roy Stogner wrote:
>
> On Mon, 4 Sep 2017, Michael Povolotskyi wrote:
>
>> using namespace libMesh;
>>
>> int main (int argc, char ** argv)
>> {
>>   {
>>     LibMeshInit init (argc, argv);
>>
>>     Mesh mesh(init.comm());
>>     MeshTools::Generation::build_cube (mesh, 5, 5, 5);
>>
>>    mesh.print_info();
>>
>>    TecplotIO mesh_output(mesh);
>>
>>     mesh_output.write("mesh.dat");
>>   }
>>
>>  return 0;
>> }
>>
>> The code is frozen at some MPI related calls inside 
>> mesh.prepare_for_use(). The libmesh was configured as:
>>
>> ./configure PETSC_DIR=/home/mpovolot/photonics/petsc/petsc-3.7.6 
>> MPIHOME=/apps/conte/openmpi/1.10.1/gcc-5.2.0 
>> PETSC_ARCH=arch-linux2-cxx-opt F77=mpif77 CC=mpicc GCC=mpicc 
>> CXX=mpiCC --disable-vtk 
>> --with-vtk-include=/depot/itap/tsaiwei/apps/vtk/6.3.0_test/include/vtk-6.3/ 
>> --with-vtk-lib=/depot/itap/tsaiwei/apps/vtk/6.3.0_test/lib 
>> --enable-tetgen --enable-triangle --enable-tecplot --disable-nemesis 
>> --disable-strict-lgpl --enable-parmesh --enable-amr 
>> --enable-shared=yes --disable-glibcxx-debugging 
>> --prefix=/home/mpovolot/photonics/libmesh/libmesh-1.2.0/built
>>
>> Could you, please, tell me what is wrong with my code?
>
> I don't see a thing wrong with it.  Are you running in dbg mode, and
> if not, would you?  Could you give us a stack trace pointing to where
> each of two processes ends up frozen?
>
> Thanks,
> ---
> Roy

Thank you,

I found that if I rebuilt everything with MPICH, instead of using 
installed openmpi, then everything works perfectly. Is libmesh supposed 
to work with openmpi? If yes, and I can recompile it again and produce 
the stack trace.

Michael.