Re: [Forces-protocol] RE: GET/SET in one msg ?

[Steven Blake <[e-mail]>]

On Thu, 2004-10-21 at 16:42, Robert Haas wrote:

> Steven Blake wrote:
> > > Do you think we should explicitely forbid the "layer-breaking" option,
> > > or could we use both ?
> > >     
> > Yes.  I consider it analogous to mixing IP addresses and ports.  It
> > requires a special case in the message parser, and it just doesn't look
> > clean.
> > 
> >   
> 
> As we don't mix layers, you could have the following scenario: MIID M
> "contains" LFBx1, LFBy1, LFBz1 (on FEs x, y, z).

I would rephrase it differently.  MIID M could be configured
individually on FEs x/y/z by the CE to alias to LFBx1, LFBy1, and LFBz1.

> But mcast ID m only spans FE x and FE y.

> So a PL message addressed to mcast ID m, and MIID M, would only act on
> LFBx1 and LFBy1.

Right.

> Or do you want to require that messages to MIID m MUST go to a mcastID
> (m or sth else) that spans at least FEs x and y ? This would be a
> layer violation too, no ?

I don't see any reason to require it in the protocol, but I also don't
see much reason to use it any other way.

> > Note that I don't see much use in needing to map a single LFB instance
> > ID (MIID) to multiple instance IDs on a single ID; i.e., I don't see the
> > need for multicast within the FE.  What is really needed is a
> > CE-assigned LFB instance ID alias that can be used on many FEs to allow
> > coordination of multicast table updates.
> > 
> >   
> I agree, but Zsolt example, if I recall well, included some level of
> intra-LFB multicasting.

I think it did.  I still don't think it is very useful, since I think
there is a better alternative for intra-FE/inter-LFB state sharing, but
it doesn't cost much to support it either.


Regards,

// Steve


_______________________________________________
Forces-protocol mailing list
[e-mail]
https://www1.ietf.org/mailman/listinfo/forces-protocol