Let's not classify protocols yet. Maybe IS-IS-TRILL is overkill and OSPFv3-&AF-&MANETextensions-&L2extensions is not. Or otherwise.
The main point is: we have to deal with v4, v6 and VLANs. Dual stack guest network may have multiple L2 links, connected via some kind of backbone link (e.g. powerline), and hooked up to a firewall somewhere. Teco Op 29 nov. 2011, om 22:52 heeft Russ White het volgende geschreven: > >> I have no doubt that if you convinced Radia about this, she could suggest >> an IS-IS-TRILL variant that would achieve it. >> >> But I'm not sure I really understand the need for it. There's no shortage >> of IPv4 address space in homenets, because they are small and in RFC 1918 >> space. So if you are routing IPv6 in a home net, why not also route IPv4? > > My point is that just as we have applications that want layer 2 > connectivity between data centers today, I can easily imagine such > applications within a home... So why not plan for a single control plane > that can route _anything_ from the very beginning. > > TRILL is way overkill for this application --go back to the original, > simple idea of just placing layer 2 addresses in a different AF in an > existing and proven routing protocol, like OSPFv3. Use some MANET > extensions to make it scale at the control plane level, and use AF's to > provide routing of anything to anything within the local domain. Assume > you won't have flooding domain boundaries, etc., and no interaction with > spanning tree, either. > > In other words --we often think about routing a particular "thing." For > homenet, it makes more sense to provide a generic control plane, and > then figure out how to provide reachability for anything you can throw > at it. IPv4? Shouldn't be an issue. IPv6? Same thing. Ethernet? A new AF > and you're done. Bluetooth extension over the network? I don't see why > it should matter --one control plane, all applications. > > Russ > _______________________________________________ > homenet mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/homenet _______________________________________________ homenet mailing list [email protected] https://www.ietf.org/mailman/listinfo/homenet
