[amarkkassery]

The python scripts don't seem to be running as well on Windows as it is on Linux.

The first thing, I noticed when looking at the parallel_computation.py script was that it looked for an environment variable called SU2_RUN and I created an environment variable and also added the path to PATH.

But now when I try to run it, it gives me the following error:

C:\Program Files (x86)\Stanford ADL\SU2\Sandbox>python ../bin/parallel_computati
on.py -p 12 -f config.cfg
--------------------------------------------------------------------------
mpirun was unable to launch the specified application as it could not access
or execute an executable:

Executable: C:\Program
Node: Workstation

while attempting to start process rank 0.
--------------------------------------------------------------------------
--------------------------------------------------------------------------
mpirun was unable to launch the specified application as it could not access
or execute an executable:

Executable: C:\Program
Node: Workstation

while attempting to start process rank 0.
--------------------------------------------------------------------------
12 total processes failed to start
'C:\Program' is not recognized as an internal or external command,
operable program or batch file.

So, it seems as though it breaks at the space between “Program Files”, so I added quotes around the run_SU2_DDC and run_SU2_CFD variables and now the error has changed to:

C:\Program Files (x86)\Stanford ADL\SU2\bin>python parallel_computation.py -p 12
-f config.cfg
--------------------------------------------------------------------------
mpirun was unable to launch the specified application as it could not access
or execute an executable:

Executable: C:\Program Files (x86)\Stanford ADL\SU2\bin\SU2_DDC config_DDC_confi
g.cfg
Node: Workstation

while attempting to start process rank 0.
--------------------------------------------------------------------------
--------------------------------------------------------------------------
mpirun was unable to launch the specified application as it could not access
or execute an executable:

Executable: C:\Program Files (x86)\Stanford ADL\SU2\bin\SU2_CFD config_CFD_confi
g.cfg
Node: Workstation

while attempting to start process rank 0.
--------------------------------------------------------------------------
12 total processes failed to start
'C:\Program' is not recognized as an internal or external command,
operable program or batch file.

I did notice that the files config_DDC_config.cfg and config_CFD_config.cfg were not created. I’m not completely sure, why.

It works fine on Linux, but just not on Windows.

[mcolonno]

Thanks for your interest in SU^2. It looks like you're running into a common problem with Python in Windows: spaces in path names. We're trying to find all these and clean them up. Please try to run from a path that does not contain any spaces and / or run directly from the executables (i.e. outside of the Python scripts). You are correct re: SU2_RUN; you need to set this per the installation documentation.

[amarkkassery]

Originally Posted by mcolonno

Thanks for your interest in SU^2. It looks like you're running into a common problem with Python in Windows: spaces in path names. We're trying to find all these and clean them up. Please try to run from a path that does not contain any spaces and / or run directly from the executables (i.e. outside of the Python scripts). You are correct re: SU2_RUN; you need to set this per the installation documentation.

Thanks Mike!

Installing it on C:\SU2 seemed to take care of that error (for now)

Would you also be able to have some kind of instruction on installing the MPICH2 libraries as just installing the executable doesn't seem to have the configuration setup correctly. Also, it doesn't seem to have the mpirun enabled. I have changed the python scripts to run mpiexec.

[mcolonno]

Glad that got you going. For instructions regarding the setup and install of MPICH2, please refer to the MPICH2 for installation guide: http://www.mpich.org/static/docs/gui...stallguide.pdf. This will guide you through installation and testing step by step.

[amarkkassery]

Originally Posted by mcolonno

Glad that got you going. For instructions regarding the setup and install of MPICH2, please refer to the MPICH2 for installation guide: http://www.mpich.org/static/docs/gui...stallguide.pdf. This will guide you through installation and testing step by step.

Thanks Mike! That helped and I verified that mpich2 was installed correctly (The version on windows doesn't come with mpirun, it only comes with mpiexec, and so far as I understand, both should work exactly in the same manner, so I changed the python script to update that). Now when I run this, this is the error that it shows:

C:\SU2\Sandbox>python "C:\SU2\bin\parallel_computation.py" -f config.cfg

-------------------------------------------------------------------------
| SU2 Suite (Domain Decomposition Code) |
-------------------------------------------------------------------------

------------------------ Physical case definition -----------------------
Input mesh file name: biplane.su2

-------------------------- Output information ---------------------------
Don't visualize the partitions.

------------------- Config file boundary information --------------------
Euler wall boundary marker(s): airfoil.
Far-field boundary marker(s): farfield.

---------------------- Read grid file information -----------------------
Three dimensional problem.
1886048 interior elements. 328995 points.
2 surface markers.
30756 boundary elements in index 0 (Marker = airfoil).
4168 boundary elements in index 1 (Marker = farfield).

----------------------- Preprocessing computations ----------------------
Identifying vertices.

--------------------------- Read FFD information ------------------------
There is no FFD box definition. Just in case, review the .su2 file

---------------------- Performing mesh partitioning ---------------------
Finished partitioning using METIS 5.0.2. (11714 edge cuts).

job aborted:
rank: node: exit code[: error message]
0: Workstation: -1073741571: process 0 exited without calling finalize
There is no geometry file (GetnZone))!
There is no geometry file (GetnZone))!
Press any key to exit...
Press any key to exit...

But when I run the DDC code without mpiexec, I get this followed by CFD_DDC.exe has stopped working error.

C:\SU2\Sandbox>C:\SU2\bin\SU2_DDC.exe config.cfg

-------------------------------------------------------------------------
| SU2 Suite (Domain Decomposition Code) |
-------------------------------------------------------------------------

------------------------ Physical case definition -----------------------
Input mesh file name: biplane.su2

-------------------------- Output information ---------------------------
Don't visualize the partitions.

------------------- Config file boundary information --------------------
Euler wall boundary marker(s): airfoil.
Far-field boundary marker(s): farfield.

---------------------- Read grid file information -----------------------
Three dimensional problem.
1886048 interior elements. 328995 points.
2 surface markers.
30756 boundary elements in index 0 (Marker = airfoil).
4168 boundary elements in index 1 (Marker = farfield).

----------------------- Preprocessing computations ----------------------
Identifying vertices.

--------------------------- Read FFD information ------------------------
There is no FFD box definition. Just in case, review the .su2 file

---------------------- Performing mesh partitioning ---------------------
Finished partitioning using METIS 5.0.2. (54928 edge cuts).

[curky]

Is anybody who can help with this? I've got stuck on the same error.

[dtucker]

I had a similar error with the "No Geometry File (GetnZone)" and looked back at a previous successfully-run config file.

The difference was the following lines I had commented out:
% --------------------------- PARTITIONING STRATEGY ---------------------------%
%
% Number of partitions of the domain
NUMBER_PART= 0
%
% Write a paraview file for each partition (NO, YES)
VISUALIZE_PART= NO

...my guess is that unless you specify otherwise, running the parallel computation python script will make the assumption that you already HAVE a partitioned mesh.

I hope that helps!
Dave