dorsal team mailing list archive

[Johannes Ring <johannr@xxxxxxxxx>]

On Mon, Dec 12, 2011 at 2:10 PM, Harish Narayanan
<harish.mlists@xxxxxxxxx> wrote:
> On 12/12/11 10:27 AM, Johannes Ring wrote:
>> On Sun, Dec 11, 2011 at 10:02 PM, Anders Logg <logg@xxxxxxxxx> wrote:
>>> On Sun, Dec 11, 2011 at 12:41:45PM +0100, Johannes Ring wrote:
>>>> On Sun, Dec 11, 2011 at 11:27 AM, Harish Narayanan
>>>> <harish.mlists@xxxxxxxxx> wrote:
>>>>> On 12/9/11 6:30 PM, Nuno Lopes wrote:
>>>>>> Lopes: Ubuntu Maverick, 32-bit:
>>>>>>
>>>>>>
>>>>>> The following optional packages could not be found:
>>>>>> -- ---------------------------------------------------
>>>>>> -- (**) PETSC
>>>>>> -- (**) SLEPC
>>>>>> -- (**) SPHYNX
>>>>>>
>>>>>>
>>>>>> Lopes: Ubuntu lucid-64b:
>>>>>>
>>>>>>
>>>>>> The following optional packages could not be found:
>>>>>> -----------------------------------------------------
>>>>>> -- (**) SPHYNX
>>>>>
>>>>> Thank you, Nuno and Zeljka for the reports.
>>>>>
>>>>> Johannes, do you know why petsc/slepc might be misbehaving on maverick
>>>>> and squeeze?
>>>>
>>>> No, I didn't have any problems with them when I tested on those
>>>> platforms. Here are my results:
>>>>
>>>> Ubuntu 10.04 LTS, 64 bit -- OK!
>>>> Ubuntu 10.10, 64 bit -- OK!
>>>> Ubuntu 11.04, 64 bit -- OK!
>>>> Ubuntu 11.10, 32 bit -- OK!
>>>> Debian sid, 32 bit -- OK!
>>>> Debian squeeze, 64 bit -- OK!
>>>>
>>>> Some comments:
>>>> - This is tested in a clean (pbuilder) environment with only a minimal
>>>> set of packages installed initially
>>>> - BOOST_DIR must be set (in addition to sourcing dolfin.conf) on
>>>> Ubuntu 10.04 and 10.10
>>>
>>> Can this be added to the Dorsal config? Same as setting the C compiler
>>> on Lion.
>>
>> Yes, I have added a note about this in the platform files now.
>
>
> Thanks, and here are some remaining issues:
>
> 1. Is it easy to define a Dorsal package file for python-netcdf and add
> it to the sid platform?

python-netcdf is available as a package in Debian so there is no need
to set Dorsal to build it for sid. I will add it to sid.platform.

> 2. Does anyone know why petsc/slepc might be broken on maverick and squeeze?

It worked fine for me in a clean environment so I don't think they are
broken, however, those who have problems with packages that cannot be
found during the DOLFIN configuration step should look in
$DOWNLOAD_PATH/src/dolfin-1.0.0/dorsal_build_dir/CMakeFiles/CMakeError.log
to see why it fails.

> 3. We need to similarly add VTK path information to the rhel/fedora
> platforms. (https://bugs.launchpad.net/dorsal/+bug/888930)
>
> Anything else?

I have been trying to get FEniCS to work correctly with the
instructions for SnowLeopard, but I currently have problems with
OpenMPI when running DOLFIN program (see below).

I'm also wondering if Dorsal should build Boost on Fedora because of
the non-working boost-math package (see
https://answers.launchpad.net/dorsal/+question/179760).

Johannes

$ python -c "import dolfin"
[buildbotmac.simula.no:49989] [NO-NAME] ORTE_ERROR_LOG: Unknown error:
-127 in file /opt/local/var/macports/build/_Volumes_work_mports_dports_science_openmpi/openmpi/work/openmpi-1.5.4/orte/mca/ess/singleton/ess_singleton_module.c
at line 231
[buildbotmac.simula.no:49989] [NO-NAME] ORTE_ERROR_LOG: Unknown error:
-127 in file /opt/local/var/macports/build/_Volumes_work_mports_dports_science_openmpi/openmpi/work/openmpi-1.5.4/orte/mca/ess/singleton/ess_singleton_module.c
at line 140
[buildbotmac.simula.no:49989] [[INVALID],INVALID] ORTE_ERROR_LOG: A
system-required executable either could not be found or was not
executable by this user in file
/opt/local/var/macports/build/_Volumes_work_mports_dports_science_openmpi/openmpi/work/openmpi-1.5.4/orte/runtime/orte_init.c
at line 128
--------------------------------------------------------------------------
It looks like orte_init failed for some reason; your parallel process is
likely to abort.  There are many reasons that a parallel process can
fail during orte_init; some of which are due to configuration or
environment problems.  This failure appears to be an internal failure;
here's some additional information (which may only be relevant to an
Open MPI developer):

  orte_ess_set_name failed
  --> Returned value A system-required executable either could not be
found or was not executable by this user (-127) instead of
ORTE_SUCCESS
--------------------------------------------------------------------------
--------------------------------------------------------------------------
It looks like MPI_INIT failed for some reason; your parallel process is
likely to abort.  There are many reasons that a parallel process can
fail during MPI_INIT; some of which are due to configuration or environment
problems.  This failure appears to be an internal failure; here's some
additional information (which may only be relevant to an Open MPI
developer):

  ompi_mpi_init: orte_init failed
  --> Returned "A system-required executable either could not be found
or was not executable by this user" (-127) instead of "Success" (0)
--------------------------------------------------------------------------
[buildbotmac.simula.no:49989] *** An error occurred in MPI_Init
[buildbotmac.simula.no:49989] *** on a NULL communicator
[buildbotmac.simula.no:49989] *** Unknown error
[buildbotmac.simula.no:49989] *** MPI_ERRORS_ARE_FATAL: your MPI job
will now abort
--------------------------------------------------------------------------
An MPI process is aborting at a time when it cannot guarantee that all
of its peer processes in the job will be killed properly.  You should
double check that everything has shut down cleanly.

  Reason:     Before MPI_INIT completed
  Local host: buildbotmac.simula.no
  PID:        49989
--------------------------------------------------------------------------