> > 1. The DNS can inject a route. Do you see problems > > with this? > > Well, "the DNS" doesn't do routes, but I assume you meant the DNS > server. Yes, there are issues: > > a) DNS server now has to implement one or more routing protocols, > which is a lot more code than, say, a stub DHCP server. > > b) What's the security model by which the router decides whether to > accept routing updates from the DNS server?
=> The same model that is used between routers in the network. I was only listing them in point form here, but the details are already in the draft. > > > 2. Using Neighbour discovery and periodic solicitations. > > Most people in the DT didn't like this option. > > > > 3. The cleanest way: extensions to MLD to allow a node > > to join an anycast group. > > These are both polling approaches, right? => Yes. Exactly the same > issues Bob > suggested apply to the co-resident DHCP approach apply here: how > tightly coupled does the thing you're polling have to be to the > service you're providing in order for the poll to tell you > whether the > service itself is available? => That's up to your implementation. The DNS is the only server using this address, so you can tie them together. Hesham -------------------------------------------------------------------- IETF IPng Working Group Mailing List IPng Home Page: http://playground.sun.com/ipng FTP archive: ftp://playground.sun.com/pub/ipng Direct all administrative requests to [EMAIL PROTECTED] --------------------------------------------------------------------
