[454514566@qq.com]

I want to Parallel compute a file in windows 64,the computer is installed :
mpich2-1.4.1p1-win-ia32.msi
SU2-3.0.000-64-bit-Parallel.exe
python 2.7
numpy and scipy
When computing a file,I enter the "python parallel_computation.py -f
turb_ONERAM6.cfg -p 2 " in cmd.exe and face the error:

the command:mpiexec -np 2 D:\Program Files\SU2\SU2_DDC config_DDC.cfg
the location: D:\
launch failed:CreateProcess<D:\Program Files\SU2\SU2_DDC config_DDC.cfg>on 'owner-PC' failed,error 2 -system can't the file
launch failed:CreateProcess<D:\Program Files\SU2\SU2_DDC config_DDC.cfg>on 'owner-PC' failed,error 2 -system can't the file
(.su2 and .cfg are located at D disk,the softare are located at D:\Program Files\SU2 and make environment variable as 'path' )

I want to the reason.Thanks

[mcolonno]

Can you please run one of the test MS MPI programs to confirm MPI is properly installed and configured? This will help eliminate possible sources when running SU^2. Thanks.

[mcolonno]

In addition, please upgrade to the latest version (3.1.0) and note that only Microsoft MPI (free add-in package) is supported, not MPICH2 for Windows which MPICH no longer supports.

[454514566@qq.com]

Originally Posted by mcolonno

In addition, please upgrade to the latest version (3.1.0) and note that only Microsoft MPI (free add-in package) is supported, not MPICH2 for Windows which MPICH no longer supports.

thank your advice,I reinstall the MS MPI and SU2 to C:,and it seem run,but I face the issue:
D:\>parallel_computation.py -f inv_NACA0012.cfg -p 2
the command: mpiexec -np 2 C:\SU2\SU2_DDC config_DDC.cfg
the location: D:\
WARNING: unrecognized option in the config. file: GRID_DEFORM_METHOD.
WARNING: unrecognized option in the config. file: GRID_DEFORM_METHOD.


-------------------------------------------------------------------------
| _____ _ _ ___ |
| / ____| | | | | |__ \ Web: su2.stanford.edu |
-------------------------------------------------------------------------
| _____ _ _ ___ |
| / ____| | | | | |__ \ Web: su2.stanford.edu |
| | (___ | | | | ) | Twitter: @su2code |
| | (___ | | | | ) | Twitter: @su2code |
| \___ \ | | | | / / Forum: www.cfd-online.com/Forums/su2/ |
| ____) | | |__| | / /_ |
| \___ \ | | | | / / Forum: www.cfd-online.com/Forums/su2/ |
| ____) | | |__| | / /_ |
| |_____/ \____/ |____| Suite (Domain Decomposition Code) |
| |_____/ \____/ |____| Suite (Domain Decomposition Code) |
| Release 2.0.10 |
-------------------------------------------------------------------------
| Release 2.0.10 |
| Stanford University Unstructured (SU2). |
| Copyright (C) 2012-2013 Aerospace Design Laboratory (ADL). |
| SU2 is distributed in the hope that it will be useful, |
| but WITHOUT ANY WARRANTY; without even the implied warranty of |
| MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
| Lesser General Public License (version 2.1) for more details. |
-------------------------------------------------------------------------

-------------------------------------------------------------------------
------------------------ Physical case definition -----------------------
Input mesh file name: mesh_NACA0012_inv.su2
| Stanford University Unstructured (SU2). |

-------------------------- Output information ---------------------------
| Copyright (C) 2012-2013 Aerospace Design Laboratory (ADL). |
Don't visualize the partitions.

------------------- Config file boundary information --------------------
| SU2 is distributed in the hope that it will be useful, |
| but WITHOUT ANY WARRANTY; without even the implied warranty of |
| MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
| Lesser General Public License (version 2.1) for more details. |
Euler wall boundary marker(s): airfoil.
Far-field boundary marker(s): farfield.
-------------------------------------------------------------------------

------------------------ Physical case definition -----------------------
Input mesh file name: mesh_NACA0012_inv.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 -----------------------
Two dimensional problem.
10216 interior elements.
10216 triangles.
5233 points.
2 surface markers.
200 boundary elements in index 0 (Marker = airfoil).
50 boundary elements in index 1 (Marker = farfield).

------------------------ Divide the numerical grid ----------------------

----------------------------- Write mesh files --------------------------

job aborted:
[ranks] message

[0] terminated

[1] process exited without calling finalize

---- error analysis -----

[1] on XingYu-PC
C:\SU2\SU2_DDC ended prematurely and may have crashed. exit code 0xc0000005

---- error analysis -----
SU2 process was terminated by signal '3'
I look forward to your answer,thank you so much!

[ceanwang]

SU2_DDC ended prematurely and may have crashed.

[economon]

Hi,

We have recently made some major improvements to the MPI implementation that will be released very soon in V3.2. Please stay tuned for updated binaries that will hopefully clear up issues.

Cheers,
Tom

[454514566@qq.com]

Originally Posted by economon

Hi,

We have recently made some major improvements to the MPI implementation that will be released very soon in V3.2. Please stay tuned for updated binaries that will hopefully clear up issues.

Cheers,
Tom

I am sorry reply late,I try to this case using the SU2 V3.1 now, and I still face the question:
D:\>parallel_computation.py -f inv_NACA0012.cfg -p 2
the command: mpiexec -np 2 C:\SU2\SU2_DDC config_DDC.cfg
the location: D:\


-------------------------------------------------------------------------
-------------------------------------------------------------------------
| _____ _ _ ___ |
| / ____| | | | | |__ \ Web: su2.stanford.edu |
| _____ _ _ ___ |
| | (___ | | | | ) | Twitter: @su2code |
| / ____| | | | | |__ \ Web: su2.stanford.edu |
| \___ \ | | | | / / Forum: www.cfd-online.com/Forums/su2/ |
| | (___ | | | | ) | Twitter: @su2code |
| ____) | | |__| | / /_ |
| \___ \ | | | | / / Forum: www.cfd-online.com/Forums/su2/ |
| |_____/ \____/ |____| Suite (Domain Decomposition Code) |
| ____) | | |__| | / /_ |
| Release 3.1.0 "eagle" |
-------------------------------------------------------------------------
| SU2, Copyright (C) 2012-2014 Aerospace Design Laboratory (ADL). |
| SU2 is distributed in the hope that it will be useful, |
| |_____/ \____/ |____| Suite (Domain Decomposition Code) |
| but WITHOUT ANY WARRANTY; without even the implied warranty of |
| Release 3.1.0 "eagle" |
| MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
-------------------------------------------------------------------------
| Lesser General Public License (version 2.1) for more details. |
| SU2, Copyright (C) 2012-2014 Aerospace Design Laboratory (ADL). |
-------------------------------------------------------------------------

------------------------ Physical case definition -----------------------
| SU2 is distributed in the hope that it will be useful, |
Input mesh file name: mesh_NACA0012_inv.su2
| but WITHOUT ANY WARRANTY; without even the implied warranty of |

-------------------------- Output information ---------------------------
| MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
Don't visualize the partitions.
| Lesser General Public License (version 2.1) for more details. |

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


------------------------ Physical case definition -----------------------
---------------------- Read grid file information -----------------------
Input mesh file name: mesh_NACA0012_inv.su2
Two dimensional problem.

5622 interior elements.
-------------------------- Output information ---------------------------
Don't visualize the partitions.

------------------- Config file boundary information --------------------
Euler wall boundary marker(s): airfoil.
Far-field boundary marker(s): farfield.
5622 triangles.
2897 points, and 0 ghost points.
2 surface markers.
137 boundary elements in index 0 (Marker = airfoil).
35 boundary elements in index 1 (Marker = farfield).

------------------------ Divide the numerical grid ----------------------

----------------------------- Write mesh files --------------------------
Mesh writing done (mesh_NACA0012_inv).

---------------------- Read and write FFD information -------------------
0 Free Form Deformation boxes.
0 Free Form Deformation nested levels.
0 Free Form Deformation boxes.
There is no FFD box definition. Just in case, review the .su2 file
0 Free Form Deformation nested levels.
There is no FFD box definition. Just in case, review the .su2 file

job aborted:
[ranks] message

[0-1] fatal error
Fatal error in MPI_Buffer_detach: Invalid argument, error stack:
MPI_Buffer_detach(135): MPI_Buffer_detach(buf=0x000000013FDF7300, size=0x000000
000000000) failed
MPI_Buffer_detach(110): Null pointer in parameter size

---- error analysis -----

[0-1] on Person-PC
mpi has detected a fatal error and aborted C:\SU2\SU2_DDC

---- error analysis -----
SU2 process was terminated by signal '4'

[economon]

Originally Posted by 454514566@qq.com

[0-1] fatal error
Fatal error in MPI_Buffer_detach: Invalid argument, error stack:
MPI_Buffer_detach(135): MPI_Buffer_detach(buf=0x000000013FDF7300, size=0x000000
000000000) failed
MPI_Buffer_detach(110): Null pointer in parameter size

I see.. we are about to release V3.2, and in this new version, we have overhauled the MPI implementation in the code. Most importantly, we have removed all MPI buffers from the code in favor of different communication types (they were only in a couple of locations, but it is an important modification). Stay on the lookout for this: we plan to release V3.2 next week.

Cheers,
Tom

[454514566@qq.com]

Originally Posted by economon

I see.. we are about to release V3.2, and in this new version, we have overhauled the MPI implementation in the code. Most importantly, we have removed all MPI buffers from the code in favor of different communication types (they were only in a couple of locations, but it is an important modification). Stay on the lookout for this: we plan to release V3.2 next week.

Cheers,
Tom

Thank you ,I am looking forward to the V3.2, I hope this is fit for the windows.