Hello, I’m not going to argue about which routing protocol to use (Juliusz remark about GRE tunnel was interesting thought), but…
See how hard it is ? Home routers vendors will have a bad enough time putting a single routing protocols in their routers, we can’t ask them to put two of them. Nevertheless, their is so much ideology beside that discussion… We are going to choose one, and what if we change our minds in 10 or 20 years ? What if a new protocol is created that fits way better homenet requirements ? Once vendors are cool with one routing protocol, we might want to allow home networks to move to a new one, or even the next version of the previous one. I think it doesn’t cost much (really) to allow flexibility. - Someone that administrate all the routers in the home should be able to use a different routing protocol than the one that homenet chose. - In the future, we should be able to move from one routing protocol to another without backward compatibility (And that might mean that transition routers will have to support two routing protocols until it becomes useless to support the previous one). So even if most will agree that supporting multiple routing protocol is a madness in the general case. It’s not that hard to ’support it’ while requiring one single routing protocol as mandatory in home networks. And whenever we want to move to another protocol, maybe in 20 years, it will allow transitioning softly. Have a nice day, Pierre Le 31 mai 2014 à 23:48, Gert Doering <g...@space.net> a écrit : > Hi, > > On Sat, May 31, 2014 at 12:17:16PM -0700, Douglas Otis wrote: >>>> On Sat, 31 May 2014, Gert Doering wrote: >>>> >>>>> So I'd keep the list of supported protocols as small as possible - and >>>>> stick to IP protocols. ISIS is great for ISP environments, but does not >>>>> nicely adapt to a unix environment where the kernel has no idea about >>>>> ISO/OSI protocols and you have to do everything via raw sockets. Which >>>>> would be a fairly typical environment for a CPE router. > [..] >> >> I could be wrong, but I don't think that was the point. There are also layer >> 2 protocols to consider. > > I quoted my original comment above. This is the context we're talking > about: protocols to be considered for routing inside the homenet. > > In *homenet*. > >> Expecting everything to be handled at IP transport layer 3 will confront >> extremely difficult security issues. >> >> Testing using a modern printer/scanner illustrated a major problem when >> devices were not restricted to link-local. Something like RBridge >> supporting PPP would provide a much safer foundation upon which to build. >> >> All networks must begin with layer 2 starting points. Selective routing >> between Rbridges would allow an ability to share data between HDCP enforced >> multi-media display devices using link-local addresses. There will be video >> cameras, baby monitors, HVAC/SCADA controls, printers and scanners offering >> media stick access, etc. These devices were never intended to directly >> interface with the Internet. These devices MUST NOT be assigned routable >> IPv4 or IPv6 addresses. Using mDNS proxy into DNS would be setting the >> stage for major security disasters. > > While this is all true, I cannot see how this related to what I said > above, and to the question of "is ISIS better suited as a link-state > protocol to transport opaque LSAs than OSPFv3?" - because *both* do that > job without "assigned routable IPv4 or IPv6 addresses" - ISIS talks over > L2 (or potentially IPv6 link-local as was mentioned), OSPFv3 talks over > IPv6 link-local. > > This thread wasn't about questioning the whole homenet architecture, but > about a specific side aspect: would additional routing protocols be considered > a plus. And I say "no", out of general reasoning, and because I consider > the particular protocol to be not very well suited for typical SoHo router > implementations. > > Gert Doering > -- NetMaster > -- > have you enabled IPv6 on something today...? > > SpaceNet AG Vorstand: Sebastian v. Bomhard > Joseph-Dollinger-Bogen 14 Aufsichtsratsvors.: A. Grundner-Culemann > D-80807 Muenchen HRB: 136055 (AG Muenchen) > Tel: +49 (0)89/32356-444 USt-IdNr.: DE813185279 > _______________________________________________ > homenet mailing list > homenet@ietf.org > https://www.ietf.org/mailman/listinfo/homenet _______________________________________________ homenet mailing list homenet@ietf.org https://www.ietf.org/mailman/listinfo/homenet
