On Mon, 2 Dec 2002, Brian Messenger wrote:
> We have a small 16 node cluster that is not passing the test cluster as
> root command. It fails on the 2 MPI tests. This is oscar 2.0 and we
> installed both lam and MPICH. Is there any trouble shooting process
> that you guys could recommend?
Can you send in the .err files that were generated?
There should probably be .out and .err files that were generated with each
failed test -- the .out corresponds the the output from stdout of the
test, and the .err file corresponds to the output on stderr.
The .err files should therefore shed some light on what happened properly.
In OSCAR 2.0, each package's tests are in their respective subdirectories
-- so look in oscar-2.0/packages/*/testing. For LAM and MPICH, there are
test_user and pbs_script.[lam|mpich] in their respective directories. If
you want to run the test manually, you can qsub the pbs_script.[lam|mpich]
scripts.
Or you have have a look in those scripts and see what the tests are doing
(they're pretty simple, actually -- just attempt to compile and run a few
MPI programs), and try to do those steps manually on your nodes (either in
PBS or not -- it probably doesn't matter for testing purposes).
--
{+} Jeff Squyres
{+} [EMAIL PROTECTED]
{+} http://www.lam-mpi.org/
-------------------------------------------------------
This SF.net email is sponsored by: Get the new Palm Tungsten T
handheld. Power & Color in a compact size!
http://ads.sourceforge.net/cgi-bin/redirect.pl?palm0002en
_______________________________________________
Oscar-users mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/oscar-users
