Sorry for not replying earlier -- yes, this should work. It's a
known issue that we have MCA parameter "contamination" if you try to
fork/exec an mpirun from within an MPI process.
On Jul 26, 2007, at 1:48 AM, Per Madsen wrote:
> Hi
>
> I have found a solution to my problem with executing a MPI program
> from within a MPI program via a system call.
>
> A related problem has been discussed in the thread: “Recursive use
> of orterun”.
>
> The problem was the setup and export of environment variables.
>
> I solved my problem by adding the following to the script starting
> the second MPI program:
>
> for i in $(env | grep OMPI_MCA |sed 's/=/ /' | awk '{print $1}')
> do
> unset $i
> done
>
> This clear the environment for the child process for all variables
> with name starting with “OMPI_NCA”
>
> Regards
> Per Madsen
> Senior scientist
>
> AARHUS UNIVERSITET / UNIVERSITY OF AARHUS
> Det Jordbrugsvidenskabelige Fakultet / Faculty of Agricultural
> Sciences
> Forskningscenter Foulum / Research Centre Foulum
> Genetik og Bioteknologi / Dept. of Genetics and Biotechnology
> Blichers Allé 20, P.O. BOX 50
> DK-8830 Tjele
>
> _______________________________________________
> users mailing list
> users_at_[hidden]
> http://www.open-mpi.org/mailman/listinfo.cgi/users
--
Jeff Squyres
Cisco Systems
|
![[Flag]](../../../../../community/mirrors/flag-us.gif){kind=small}