3. apr. 2014 kl. 18:26 skrev Imran Ali <[email protected]>:
>
> 3. apr. 2014 kl. 12:31 skrev Imran Ali <[email protected]>:
>
>> On 2014-04-03 11:45, Johannes Ring wrote:
>>> On Thu, Apr 3, 2014 at 10:33 AM, Imran Ali
>>> <[email protected]> wrote:
>>>> Using Redhat Enterprise 6, I am installing following packages through
>>>> dorsal
>>>> :
>>>> bzip2 boost cmake pcre swig glib2 libffi git vtkdata vtk cgal eigen mpfr
>>>> ply
>>>> python numpy matplotlib scientificpython cython scipy ipython mpi4py hdf5
>>>> h5py parmetis scotch blas lapack trilinos_intel petsc petsc4py slepc fiat
>>>> ferari ufc ufl ffc instant dolfin
>>> Note that the order you specify these packages is important for
>>> Dorsal. If the above list is what you have used, then for instance
>>> 'vtk' and 'ply' would be installed with system python (which I assume
>>> is version 2.6) because they come before 'python', while for instance
>>> 'numpy' will use the python built by Dorsal (which is version 2.7).
>>> You should therefor put 'python' early in the list to make sure all
>>> packages that depends on python are built with the same python.
>>>> I end up with two versions of python in my Work directory. I am guessing
>>>> that if I did not include the python package, I would only have 2.6. I
>>>> notice that both versions are used interchangeably when working with FEniCS
>>>> applications. Will this create any problems ?
>>> Yes, most likely.
>>> Johannes
>>
>> I just wanted to add the following warning I get when importing dolfin :
>>
>> import dolfin
>> /uio/hume/student-u29/imranal/Work/FEniCS/lib64/python2.6/site-packages/petsc4py/lib/__init__.py:64:
>> RuntimeWarning: compiletime version 2.6 of module 'petsc4py.PETSc' does not
>> match runtime version 2.7
>> module = imp.load_module(fullname, fo, fn, stuff)
>> /uio/hume/student-u29/imranal/Work/FEniCS/lib64/python2.6/site-packages/petsc4py/lib/PETSc.so:4:
>> RuntimeWarning: compiletime version 2.6 of module 'libpetsc4py' does not
>> match runtime version 2.7
>>
>> Thanks Johannes. I will try to fix my fenics install by rearranging those
>> packages.
>>
>> Imran
>
> I am afraid that simply worsened things. Installing Python first simply made
> dolfin unsuable. Importing dolfin generated error : No module named common.
>
> Reinstalling python2.7 only, I reverted back to the warning above. However,
> now PETSc crashes the ipython session :
>
>
> RuntimeWarning: compiletime version 2.6 of module 'petsc4py.PETSc' does not
> match runtime version 2.7
> module = imp.load_module(fullname, fo, fn, stuff)
> /uio/hume/student-u29/imranal/Work/FEniCS/lib64/python2.6/site-packages/petsc4py/lib/PETSc.so:4:
> RuntimeWarning: compiletime version 2.6 of module 'libpetsc4py' does not
> match runtime version 2.7
>
> [0]PETSC ERROR:
> ------------------------------------------------------------------------
> [0]PETSC ERROR: Caught signal number 11 SEGV: Segmentation Violation,
> probably memory access out of range
> [0]PETSC ERROR: Try option -start_in_debugger or -on_error_attach_debugger
> [0]PETSC ERROR: or
> seehttp://www.mcs.anl.gov/petsc/documentation/faq.html#valgrind[0]PETSC
> ERROR: or tryhttp://valgrind.org on GNU/linux and Apple Mac OS X to find
> memory corruption errors
> [0]PETSC ERROR: configure using --with-debugging=yes, recompile, link, and run
> [0]PETSC ERROR: to get more information on the crash.
> --------------------------------------------------------------------------
> MPI_ABORT was invoked on rank 0 in communicator MPI_COMM_WORLD
> with errorcode 59.
>
> 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.
> --------------------------------------------------------------------------
>
Reinstalling PETSc and PETSc4py removed the error. Now I am not receiving the
warning either, which I was originally.
Imran
_______________________________________________
fenics-support mailing list
[email protected]
http://fenicsproject.org/mailman/listinfo/fenics-support
