[CIG-SHORT] problem with Multiprocessing

[Eric Heien]

Hi Alireza,

Thanks for letting us know about this.  I just fixed the bug and updated the repository.  If you do "git pull” and recompile it should work now.

Best,

-Eric

On Sep 12, 2014, at 1:15 AM, Alireza Kh <civil.alireza at gmail.com> wrote:

> I am trying to install and test Virtual California for the first time on Ubuntu 14.04. According to the User Manual, I got the essential and optional requirements (such as openMP, SWIG…) to compile VC and deploy it across multiple processors. Fortunately, all tests passed (as shown in the attached file). However, I have a problem with writing output files in HDF5 Format, when I use multiple processors. For the first attempt to fix the problem, I thought it comes from improper installation of HDF5-dev, and changed the output file format to TXT. Unfortunately, the results from an MPI simulation were not identical that from a single processor simulation (the attached sample). I did my best to find the wrong thing(s) in installation procedure, but I could not find causes of error “Segmentation fault (Signal 11)”. Would you please help me about this problem?
> 
> 
> sincerely yours,
> Alireza
> 
> <Problem log.docx><sample.tar.gz>_______________________________________________
> CIG-SHORT mailing list
> CIG-SHORT at geodynamics.org
> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-short