Re: how to learn the peerID that has published a PipeAdvertisement?

Is (should) an advertisement really peer specific? Is it possible that
an advertisement is more of a group level adv than a peer level adv
(like an adv for a service or a pipe for a group service) and therefore
should not be signed or more assigned to a specific peer?

At first I thought that the PeerID should have been in the PipeAdv, but
then it would makes sense that the pipe advertisement in a redundant
service is the same for several peers running the group service.

Not sure...

Fred

--- Mohamed Abdelaziz <[email protected]> wrote:

> The most viable, and definitive way to determine the origin of an 
> advertisement is to sign it.  It would be a great contribution to the
> 
> platform.
> 
> Any takers?
> 
> Mohamed
> 
> -- 
> http://blogs.sun.com/roller/page/hamada
> http://weblogs.java.net/blog/hamada
> 
> 
> 
> Mike Cumings wrote:
> 
> >John,
> >
> >I dont know of any direct way to do this.  Pipes themselves are
> >coupled more tightly to the peer group than to any one particular
> >peer.  As a result, you could use the PipeID to work backwards to
> >the PeerGroupID, but you wouldnt be able to get an ID of the peer.
> >
> >The discovery service only (sometimes) gives you the responding
> >peer's advertisement, though since requests may be forwarded I
> >dont think this can be used.
> >
> >From what I can tell, the shared resource distributed index (SRDI)
> >data store does not store the publishing peer's info along with the
> >resource being indexed - it only stores the data being indexed.
> >Not 100% on this one though...
> >
> >Mike
> >
> >
> >On 9/6/05, John D. Blair <[email protected]> wrote:
> >  
> >
> >>How can I determine the peerID of the peer that has published a
> >>PipeAdvertisement?  Further, can this information be learned for
> any
> >>advertisement?
> >>
> >>thanks!
> >>  -john.
> >>
> >>    
> >>
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [email protected]
> For additional commands, e-mail: [email protected]
> 
>