On Sep 3, 2004, at 6:10 PM, Bernard Li wrote:
> What would be the best way to prevent the SESSION_SUFFIX from being
> used
> then? Would it simply be modifying the lam-conf.lamd and commenting
> out
> $session_suffix?
>
> I suspect the problem I am encountering has to do with the application
> not being able to attach to the same session (what you described).
> Also, is it possible to tell mpirun which session to attach to? I
> guess
> I could use the LAM_MPI_SESSION_SUFFIX environment variable?
By "SESSION_SUFFIX", I was speaking shorthand for
"LAM_MPI_SESSION_SUFFIX". :-)
It's been a while since I've looked at the SGE scripts, but don't they
use LAM_MPI_SESSION_SUFFIX? I'm thinking that you have to use it for
the SGE integration... The only way that mpirun knows what session to
attach to is through the LAM_MPI_SESSION_SUFFIX and/or a handful of
other environment variables (i.e., ones specific to each batch system).
IMHO, the real solution here would be to fix Java (if, indeed, it is
not passing the environment through to the MPI processes). ;-) This
may well be against the "sandbox" philosophy of a java runtime, though
-- remember my disclaimer that I have no java experience. :-)
--
{+} Jeff Squyres
{+} jsquyres_at_[hidden]
{+} http://www.lam-mpi.org/
|
