On Oct 28, 2010, at 12:46 AM, Fred Baker wrote: >>> If it really has to be a literal, which I already pointed out has some >>> brain-deadness related to IPv4 and IPv6 routing (the fact that you and I >>> both have an IPv[46] address doesn't mean that the network has a route that >>> connects them, even if the addresses are global). But that is not is not >>> its *own* address; it is the address of a neighbor. >> >> The referral doesn't have to be a literal in all cases, but a DNS name is >> not a general solution that applications developers can use. > > Select an alternative. If a DNS name, which encompasses all reasonable IP[46] > addresses, is unreasonable, and a single literal that may or may not have a > functional route associated with it from its peer's perspective is > unreasonable, what is reasonable? I haven't heard you suggest an alternative > that predictably works.
Given that DNS names don't even begin to solve the problem, and there's not currently support in the network for any identifier that will work, the most workable approach for now requires applications to pass IP addresses to peers in referrals (perhaps with some additional information) and for those peers to make heuristic guesses about which addresses to try first. Granted that's not a very good solution, but it's way better than trying to prevent apps from making those decisions. As for an eventual good solution: Building a fast, secure, reliable split ID/LOC system strikes me as approximately as difficult as building a routing system that scales to 10^7 prefixes - because the likely solution to both seems to involve a very highly replicated, reliable, fast, and secure distributed database that everyone is willing to trust. It doesn't mean we shouldn't try to solve both problems but that the answers aren't likely to be simple enough to be communicated in an email for which you're only willing to wait 48 hours. >>> If you are adamant that the web/sip/whatever referral can't be a DNS name, >>> will you allow the referring host to look it up in DNS? As noted, DNS will >>> have the external addresses of any system it has a name for. >> >> see above. A DNS name is not a general solution. > > I repeat. You have given a lot of crap about how you don't like solutions > that have been offered. I have offered solutions. I don't recall you offering any solutions. You keep saying DNS but I've explained why DNS doesn't even begin to work. Do you need more detail? I thought we'd been over this many times before... > Pick one or suggest one. seems like you're insisting on simple answers to things that inherently aren't simple. Keith _______________________________________________ nat66 mailing list [email protected] https://www.ietf.org/mailman/listinfo/nat66
