Ok, thank you. The configuration now seems to have the optional include directories:
libmesh_optional_INCLUDES........ : -I/usr/include/eigen3 -I/usr/include -I/home/tobias/PETSc/petsc-install//include -I/home/tobias/PETSc/petsc-install//arch-linux2-c-debug/include -I/home/tobias/PETSc/petsc-install/include libmesh_optional_LIBS............ : -L/lib -lhdf5 -Wl,-rpath,/lib -lz -ltbb -ltbbmalloc -Wl,-rpath,/home/tobias/PETSc/petsc-install//lib -L/home/tobias/PETSc/petsc-install//lib -lpetsc -L/home/tobias/PETSc/petsc-install/lib -lflapack -lfblas -lX11 -lpthread -lssl -lcrypto -lmpifort -lgfortran -lquadmath -lm -lmpicxx -lstdc++ -Wl,-rpath,/home/tobias/PETSc/petsc-install/lib -lmpi -lgcc_s -ldl -Wl,-rpath,/usr/lib -L/usr/lib -L/usr/lib/gcc/x86_64-redhat-linux/4.8.2 -Wl,-rpath,/usr/lib/gcc/x86_64-redhat-linux/4.8.2 However, this path looks strange to me (it has a double // ): -I/home/tobias/PETSc/petsc-install//include Is this somehow related to the fact that "arch-linux2-c-debug" is in "/home/tobias/PETSc/petsc-3.6.2" rather than in "/home/tobias/PETSc/petsc-install/" ? ________________________________ Von: John Peterson <jwpeter...@gmail.com> Gesendet: Donnerstag, 5. November 2015 15:19 An: tschroeder Cc: libmesh-users Betreff: Re: [Libmesh-users] Neumann boundary problem for Laplace equation On Thu, Nov 5, 2015 at 1:17 PM, tschroeder <tschroe...@daad-alumni.de<mailto:tschroe...@daad-alumni.de>> wrote: Could the error be related to this statement in my PETSc install: I think the problem is that you need to run the command *exactly* as the PETSc installation process tells you to do. That is: make PETSC_DIR=/home/tobias/PETSc/petsc-3.6.2 PETSC_ARCH=arch-linux2-c-debug install not simply "make install". -- John ------------------------------------------------------------------------------ _______________________________________________ Libmesh-users mailing list Libmesh-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/libmesh-users
