Capturing return values from concurrent mpiexecs

[Martin Schafföner]

On Tuesday 08 November 2005 20:15, Pete Wyckoff wrote:

> What you describe wanting is how I thought it worked now, but
> testing shows it does not.  Did it ever work in an earlier version?
> Are you using 0.80 now?  I'll play with it and see if I can

Don't know if it worked before; I only started using mpiexec now (0.80) to 
replace all rsh-related stuff on our nodes.

> I'd imagine that "mpiexec -server" would always return 0 unless it
> died in some unnatural way---it would never return the exit status
> of tasks it started for any of its clients.

If I send SIGTERM to "mpiexec -server", it returns 1.

> The combined return value of a parallel task is always just the exit
> status of task #0, although there will be warning lines on stderr to
> report the non-zero exit statuses of other tasks.  I couldn't come
> up with any better way of reporting the array of exit statuses.  Any
> suggestions?

Well, I guess only returning the exit status of task #0 and warning about 
others is fine, I think. Only if I start, say, 8 concurrent "mpiexec -n x" 
while "mpiexec -server" is running I would like to get 8 return values.

I have also been tinkering with the source code, adding trace stuff here and 
there, but have not found the cause of the failure.

Regards,
-- 
Martin Schafföner

Cognitive Systems Group, Institute of Electronics, Signal Processing and 
Communication Technologies, Department of Electrical Engineering, 
Otto-von-Guericke University Magdeburg
Phone: +49 391 6720063