Hi Pekka, I think i was too focussed in the site multihoming case :-)
However, in the particular case of multihoming, the accepted source prefix is likely to be only one, i think. So in this case it would be useful to include it to allow the host to discover the appropriate prefix to be inlcuded in the source address. So, for the multihoming case, this would be IMHO pretty useful and it would worth the effort, i think. I agree that a MAY would do it, so that exit routers at multihomed sites can be configured to support this and hosts within the multihomed site are upgraded to understand the message regards, marcelo > -----Mensaje original----- > De: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] nombre de Pekka > Savola > Enviado el: viernes, 30 de enero de 2004 15:35 > Para: marcelo bagnulo > CC: [EMAIL PROTECTED]; [EMAIL PROTECTED] > Asunto: RE: ICMPv6: New destination unreachable codes > > > On Fri, 30 Jan 2004, marcelo bagnulo wrote: > > I mean, in the case that the packet is discarded becuase the > source address > > is not compatible with the ingress filtering, the router that > discards the > > packets probably knows which are the accepted prefixes, so it would be > > interesting to use the message error to inform the host about > the correct > > prefix to use for this particular destination. > > In general, the router does know the accepted prefixes, but the list > might be even hundreds of prefixes long, not just one prefix. > > So, in general, embedding this information is not possible or > feasible. The case of just one accepted prefix is special. Different > question is whether it makes sense to add this information. In any > case it could not be stronger than a MAY (IMHO), so you couldn't > depend anyone adding it, so you'd have to devise the other mechanisms > anyway. > > So, it seems a bit like adding the information is not worth the effort > for the special case. > > -- > Pekka Savola "You each name yourselves king, yet the > Netcore Oy kingdom bleeds." > Systems. Networks. Security. -- George R.R. Martin: A Clash of Kings > > > -------------------------------------------------------------------- > IETF IPv6 working group mailing list > [EMAIL PROTECTED] > Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 > -------------------------------------------------------------------- > -------------------------------------------------------------------- IETF IPv6 working group mailing list [EMAIL PROTECTED] Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
