Juliusz - If you were present at the IS-IS-WG meeting in Honolulu (if not consult the minutes and/or read the list archives) you may remember that there was a good deal of discussion about this problem in the context of http://www.ietf.org/id/draft-lamparter-isis-reachability-critical-subtlvs-00.txt.
The strong recommendation of myself and others was to use a reserved MTID for Homenet networks. In such a case even if a non-Homenet IS-IS router formed an adjacency with a Homenet IS-IS router they would never see each other in the same topology so neither would be part of the topology specific SPT and neither would try to forward anything via the other. So this problem is easily addressed... Les > -----Original Message----- > From: homenet [mailto:[email protected]] On Behalf Of Juliusz > Chroboczek > Sent: Monday, March 09, 2015 1:38 PM > To: Christian Hopps > Cc: Ray Hunter; [email protected]; Ted Lemon > Subject: Re: [homenet] Stub networks > > >> Brian, ever the pessimist, expects things to go wrong whenever they > >> can. I happen to agree with Brian. > >> > >> (Which, by the way, is the reason why I think that the way networks > >> containing both IS-IS and Homenet IS-IS will silently break is a > >> major f*ck up. Section 6.3. Yeah, I was too tired to argue that > >> point.) > > > The user would have to go out of their way to configure their > > non-homenet IS-IS router specifically to interoperate with the homenet > > IS-IS. How would they know how to do this other than to be familiar > > with the other requirements? It’s not like they would be able to > > accidentally drop a non-homenet IS-IS router into their network and > > have it start silently failing. > > It's 2017. There's an animated discussion on Stackoverflow about how to > install IS-IS on an ordinary PC in order to get it to interoperate with > Homenet. > There are some people who point out that the Homenet variant of IS-IS is > not interoperable with stock IS-IS, but they get voted down. > ``Haters will hate, but I've been running it that way for three days and it > works just fine''. > > I'm not asking for full interoperability, I just wish source-specific IS-IS > would > refuse to establish neighbour relationships with non-specific IS-IS. We'll > pay > for that mistake at some point. > > Ted once wrote that we cannot prevent people from being stupid. Ted is of > course right, but I still think that we should strive to minimise the > consequences of human stupidity. > > -- Juliusz > > _______________________________________________ > homenet mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/homenet _______________________________________________ homenet mailing list [email protected] https://www.ietf.org/mailman/listinfo/homenet
