OpenMPI fail with Bluecape port of OF2.1 (I am doing it wrong).
 

So I have an operating setup of Bluecapes port of OF2.1 running just fine including running in parallel on Windows 7 64.
So I'm trying to setup another machine in exactly the same way, except clearly I am screwing something up as I get the error message below in the log files when trying to run a job in parallel.

BTW the Windows firewall is totally disabled on this machine, so that should not be an issue, I believe I've followed the step in the user documentation, but clearly I am doing this wrong. Its frustrating as I've been through this once and solved it already and from memory its was an easy thing to do.

AQny help is appreciated.


C:\BLUECF~1.1\msys\home\ofuser>mpirun -n 4 -x HOME -x PATH -x USERNAME -x WM_PROJECT_DIR -x WM_PROJECT_INST_DIR -x WM_OPTIONS -x FOAM_LIBBIN -x FOAM_APPBIN -x FOAM_USER_APPBIN -x MPI_BUFFER_SIZE 5131088.bat
[WIN-UEDH712LCQI:01756] [[2578,0],0] ORTE_ERROR_LOG: Error in file ..\..\..\openmpi-1.5.3\orte\mca\ess\hnp\ess_hnp_module.c at line 214
[WIN-UEDH712LCQI:01756] [[2578,0],0] ORTE_ERROR_LOG: Error in file ..\..\..\openmpi-1.5.3\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_rml_base_select failed
--> Returned value Error (-1) instead of ORTE_SUCCESS
--------------------------------------------------------------------------
[WIN-UEDH712LCQI:01756] [[2578,0],0] ORTE_ERROR_LOG: Error in file ..\..\..\..\..\openmpi-1.5.3\orte\tools\orterun\orterun.c at line 572
--------------------------------------------------------------------------
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 Error (-1) instead of ORTE_SUCCESS
--------------------------------------------------------------------------

Hi Doug68,

Mmm... not much information to work with, although from the first command line, it looks like I've seen some similar issue before. So, a few questions:

1. In this second machine, did you use the blueCFD installer or did you copy the installation from the first machine?
   The reason for asking this is because the installer will at least try to install the MSVC 2010 runtime that Open-MPI needs.
2. Which command did you use to launch the parallel application? gompi or foamJob?
3. From the command line, it looks like you ran one of the commands mentioned before directly from the home folder. This could happen due to either one of two reasons:
   • If you did run from the home folder ("home/ofuser" to be exact).
   • If you ran the command from another drive letter that isn't "C:". If this is the case, then this is a known bug that is going to be fixed in the next version of blueCFD.

Since I suspect it's the last issue, I'll send you an email with the fixed "gompi.bat" that is going to be release in the next version of blueCFD.

Best regards,
Bruno

Sorry, all fixed now went back tot he beginning and followed the instructions properly ;)