[CIG-SHORT] Problems compiling pylith

Brad Aagaard baagaard at usgs.gov
Thu Jun 19 10:01:52 PDT 2008


Fabrizio-

You can't mix the repository version of PyLith with the posted PETSc 
tarball. They are incompatible. You need either the posted versions of 
PETSc dev and PyLith or PyLith from the repository and PETSc dev from 
the PETSc repository. Both are constantly evolving so you need snapshots 
from the same point in time. If the posted sources do not work, please 
send us the error messages (config.log and create a make.log). If there 
is an error in the source tarball, we can correct it. I believe at least 
one other user was able to successfully build PyLith using the posted 
source tarballs.

Brad



Fabrizio Pistagna wrote:
> Hi all,
> i must install pylith on a Scientific Linux CERN SLC release 4.6 
> (Beryllium) 2.6.9-55.0.6.EL.cern1smp kernel on x86_64 architecture,
> boosting four Dual-Core AMD Opteron(tm) Processor 2218.
> 
> I followed the "installation from source" chapter from official pylith 
> book, using MPI implementation mpich-1.2.7 for gcc4 compiler.
> .
> Here is my configuration in order of installation:
> - Python2.5.2
> - last petsc-dev from cig site
> - nemesis-1.0
> - proj-4.6.0
> - numpy-1.1.0
> - FIAT-0.3.4
> - Pyrex-0.9.8.4
> - pyrexembed-0.1.1
> - spatialdata-0.3.1
> - pylith got from svn
> 
> I use the last svn trunk revision because of the version 1.1.2 didn't 
> compile during make operation, due to a reference to an inexisting
> member named "restrict" at line 270-271 of source 
> files bc/AbsorbingDampers.cc:
> 
> bc/AbsorbingDampers.cc:270: error: 'class ALE::Mesh' has no member named 
> 'restrict'
> 
> All packages were installed with a --prefix custom path, setting right 
> variable for PETSC_ARCH and PETSC_DIR.
> All gone done, but when i run pylith from terminal, it executes and 
> blocks itself when it try to imports petsc module, here is the stdout:
> 
> $ pylith
> Traceback (most recent call last):
>   File "<string>", line 1, in <module>
>   File 
> "/home/fpistagna/B/packages/lib/python2.5/site-packages/pythia-0.8.1.6-py2.5.egg/mpi/__init__.py", 
> line 42, in mpistart
>     kwds = kwds)
>   File 
> "/home/fpistagna/B/packages/lib/python2.5/site-packages/pythia-0.8.1.6-py2.5.egg/pyre/applications/__init__.py", 
> line 41, in start
>     shell.run(**kwds)
>   File 
> "/home/fpistagna/B/packages/lib/python2.5/site-packages/pythia-0.8.1.6-py2.5.egg/pyre/applications/Shell.py", 
> line 142, in run
>     method(*args, **kwds)
>   File 
> "/home/fpistagna/B/packages/lib/python2.5/site-packages/pythia-0.8.1.6-py2.5.egg/pyre/applications/SuperScript.py", 
> line 49, in execute
>     self.main(*args, **kwds)
>   File 
> "/home/fpistagna/B/packages/lib/python2.5/site-packages/pythia-0.8.1.6-py2.5.egg/pyre/applications/SuperScript.py", 
> line 58, in main
>     self.runSubscript(*args, **kwds)
>   File 
> "/home/fpistagna/B/packages/lib/python2.5/site-packages/pythia-0.8.1.6-py2.5.egg/pyre/applications/SuperScript.py", 
> line 62, in runSubscript
>     self.subscript.run(*args, **kwds)
>   File 
> "/home/fpistagna/B/packages/lib/python2.5/site-packages/pythia-0.8.1.6-py2.5.egg/pyre/applications/Application.py", 
> line 42, in run
>     shell.run(*args, **kwds)
>   File 
> "/home/fpistagna/B/packages/lib/python2.5/site-packages/pythia-0.8.1.6-py2.5.egg/pyre/applications/Shell.py", 
> line 142, in run
>     method(*args, **kwds)
>   File 
> "/home/fpistagna/B/packages/lib/python2.5/site-packages/pythia-0.8.1.6-py2.5.egg/mpi/Application.py", 
> line 118, in _onComputeNodes
>     self.onComputeNodes(*args, **kwds)
>   File 
> "/home/fpistagna/B/packages/lib/python2.5/site-packages/pythia-0.8.1.6-py2.5.egg/mpi/Application.py", 
> line 124, in onComputeNodes
>     self.main(*args, **kwds)
>   File "/home/fpistagna/B/files/pylith-1.1.2/pylith/PyLithApp.py", line 
> 79, in main
>     self.petsc.initialize()
>   File 
> "/home/fpistagna/B/files/pylith-1.1.2/pylith/utils/PetscManager.py", 
> line 48, in initialize
>     import pylith.utils.petsc as bindings
> ImportError: No module named petsc
> --pyre-start: mpirun: exit 1
> 
> Any ideas?
> Thanks in advance
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> CIG-SHORT mailing list
> CIG-SHORT at geodynamics.org
> http://geodynamics.org/cgi-bin/mailman/listinfo/cig-short



More information about the CIG-SHORT mailing list