[CIG-SHORT] pylith-1.8.0

[Brad Aagaard]

On 12/15/2017 12:04 PM, Ehsan Haghighat wrote:
> Hi Brad,
> 
> I did so. There was some changes that i needed to apply. But this has something to do with underlying evaluations in petsc. If remove certain faults, it runs just fine. It has problems forming the contact elements for the same faults that used to work with 1.8.0. Do you think I can recompile 2.1.0 with 1.8.0 petsc ?

Did you mark buried edges of the faults when generating the mesh and use 
the "edge" fault property to tell PyLith which vertices are on the 
buried edges?

No, you cannot use PyLith v2.1.0 with PETSc from PyLith v1.8.0. There 
are changes to the PETSc functions etc that make them incompatible. 
Additionally, PyLith v2.1.0 uses DMPLex() for storing the finite-element 
mesh and it did not exist in PETSc for PyLith v1.8.0.

Regards,
Brad


> 
> Thanks,
> Ehsan
>> On Dec 15, 2017, at 1:59 PM, Brad Aagaard <baagaard at usgs.gov> wrote:
>>
>> Ehsan,
>>
>> If you have not done so already, I strongly recommend reading the PyLith CHANGES file https://github.com/geodynamics/pylith/blob/master/CHANGES to see all of the changes required to update .cfg files from PyLith 1.8 to the current version (2.2.1). A lot has changed in 5 years and some settings will need to be updated.
>>
>> Regards,
>> Brad
>>
>>
>>> On 12/15/2017 10:37 AM, Ehsan Haghighat wrote:
>>> Hi Matt,
>>> I am working on a 3D mesh with faults. This model is working fine with 1.8.0. However, it does not work with 2.1.0. They 2.1.0 is installed from the installer 2.1.0-3, so there should not be any inconsistencies, i assume, as I did not modify the installer at all.
>>> Thanks,
>>> Ehsan
>>>> On Dec 15, 2017, at 1:03 PM, Matthew Knepley <knepley at rice.edu <mailto:knepley at rice.edu>> wrote:
>>>>
>>>> On Fri, Dec 15, 2017 at 12:45 PM, Ehsan Haghighat <ehsanh at mit.edu <mailto:ehsanh at mit.edu>> wrote:
>>>>
>>>>     Hello Brad,
>>>>
>>>>     I cannot find the pylith-1.8.0 installer in the github anymore. I
>>>>     know this is an old version, but we need access to this since we
>>>>     cannot run some old projects in current version and not much time
>>>>     to investigate the reason. Currently I am using the 2.1.0 version.
>>>>     One problem that I am getting for instance:
>>>>
>>>>     [0]PETSC ERROR: --------------------- Error Message
>>>>     --------------------------------------------------------------
>>>>     [0]PETSC ERROR: Invalid argument
>>>>     [0]PETSC ERROR: Faces should separate only two cells, not 4
>>>>
>>>>
>>>> I assume this is a 2D Quad mesh. It is not "interpolated" meaning the faces have not been created, and you only have vertices and cells. It sounds
>>>> like you are mismatching versions of PyLith and PETSc.
>>>>
>>>>    Thanks,
>>>>
>>>>      Matt
>>>>
>>>>     [0]PETSC ERROR: See
>>>>     http://www.mcs.anl.gov/petsc/documentation/faq.html
>>>>     <http://www.mcs.anl.gov/petsc/documentation/faq.html> for trouble
>>>>     shooting.
>>>>     [0]PETSC ERROR: Petsc Development GIT revision:
>>>>     v3.5.3-2251-gb193602  GIT Date: 2015-02-19 19:20:25 -0600
>>>>     [0]PETSC ERROR:
>>>>     /nobackup1/ehsanhaghighat/dev/GRS-2.1.0/pylith/bin/mpinemesis on a
>>>>     arch-pylith named eofe5 by ehsanhaghighat Fri Dec 15 12:20:10 2017
>>>>
>>>>
>>>>     Can you please let me know if there is a chance I can get the
>>>>     installer for the 1.8.0?
>>>>
>>>>     Thanks,
>>>>     Ehsan
>>>>
>>>>
>>>>     _______________________________________________
>>>>     CIG-SHORT mailing list
>>>>     CIG-SHORT at geodynamics.org <mailto:CIG-SHORT at geodynamics.org>
>>>>     http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-short
>>>>     <http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-short>
>>>>
>>>>
>>>> _______________________________________________
>>>> CIG-SHORT mailing list
>>>> CIG-SHORT at geodynamics.org <mailto:CIG-SHORT at geodynamics.org>
>>>> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-short
>>
>> _______________________________________________
>> CIG-SHORT mailing list
>> CIG-SHORT at geodynamics.org
>> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-short
> _______________________________________________
> CIG-SHORT mailing list
> CIG-SHORT at geodynamics.org
> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-short
>