Re: TM interface errors

Jeff Squyres (jsquyres) wrote:
> Did you ever get an answer from this on the Torque users list?
>
> (per my mail from 8 Apr, Open MPI doesn't hold open the TM connection)
>   
> > -----Original Message-----
> > From: Prakash Velayutham [mailto:velayups@xxxxxxxxxxxx] 
> > Sent: Saturday, April 01, 2006 11:29 AM
> > To: David Golden
> > Cc: torqueusers@xxxxxxxxxxxxxxxx; Jeff Squyres
> > Subject: Re: [torqueusers] TM interface errors
> >
> > David Golden wrote:
> >     
> > > On 2006-03-31 13:01:47 -0500, Prakash Velayutham wrote:
> > >   
> > >       
> > > > Hi,
> > > >
> > > > I have a minimal MPI program to test the TM interface and
> > > > strangely I seem to get errors during tm_init call.
> > > > Could someone explain what could be wrong? Here is the MPI code:
> > > >
> > > >         
> > > Haven't even looked at the code; but:
> > > AFAIK there's a 1-client limit in current TM API 
> > > (see torque planned changes list on wiki). 
> > > openmpi's run time environment abstraction layer
> > > (openrte) would be that client, your user code probably
> > > can't be that client.
> > >       
> > So, are you saying that from inside of an Open MPI code I 
> > cannot call TM 
> > libraries?
> > I am copying Jeff Squyres from Open MPI team here, just to 
> > get his ideas.
> >
> > Thanks,
> > Prakash
Hi Jeff,

Nope. No replies yet. People here seem to hold the idea that TM module 
in Open MPI captures the single TM connection available. But there are 
no replies on why the same works from Mother Superior, but does not work 
from MOMs alone (with or without the TM_MULTIPLE_CONNS patch).


Thanks for asking,
Prakash