Hi, On Fri, May 03, 2002 at 08:19:57PM -0700, Michel Py wrote:
It appears to me that if you want to design a multihomed DNS chain, there could be some value in knowing what multihoming is actually going to be; and it's not going to be leaking PA prefixes in the DFZ in any scenario considered today.
Maybe I should be more precise on that. We're talking for example about the address space to be used by the RIPE NCCs DNS servers, doing reverse delegation for 6.0.1.0.0.2.ip6.{int,arpa}. The NCC is multi-homed to about 90 ISPs today, many of which are now starting to offer IPv6 services, natively delivered over the AMS-IX exchange. People want to see this service so that they can do v6-only stuff, without the need to go to ns.ripe.net's IPv4 address. 6to4 is not an answer in this scenario, neither are "use a /48 from each of the upstreams" nor "get a globally unique prefix" (because that network is *not* special enough to warrant an exception from "end sites do not get an allocation") Which leaves "leak more-specific", which would solve that problem quite well. [..]
The kind of partial redundancy you would get by leaking prefixes has been solved a year ago.
Maybe I am again overlooking something here. To my knowledge, this is still the recommended way to do BGP multihoming in IPv4 land (unless other multihoming approaches like double-IP/DNS balancing, or multihome-to-one-upstream can be used). IPv6 is not (yet) in any way different concerning routing, except that renumbering will be easier in many cases, so there are stronger arguments for "PA leaking" vs. "using PI and keep that forever". Still the basic issue is unchanged - some people do not want to rely on just one upstream, and for others (like the NCC) it doesn't make sense to confine their service to just one upstream. So what exactly are you talking about? Any implementations out there that I could look at?
What we are currently discussing are specific cases of catastrophic multiple-ISP failures.
In which, depending on the setup, all bets are open anyway...
Whining about the lack of a v6 multihoming solution is not going to bring it to you faster. I am not the one that stalled the process. Have you considered participating in its development?
I'm doing my best to help developing policies and techniques that follow the pragmatic approach of "make it work today and avoid doing irrevocable damage in the progress". Gert Doering -- NetMaster -- Total number of prefixes smaller than registry allocations: 45077 (47584) SpaceNet AG Mail: netmaster@Space.Net Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0 80807 Muenchen Fax : +49-89-32356-299