[ppml] IPv6 flawed?

On 9/13/07, mcr at xdsinc.net <mcr at xdsinc.net> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1

>   In IPv6, it is strictly not uncommon for a host to have multiple IP
> addresses.  So, you don't renumber the hosts or use NAT. You just let
> them have an IP address from your ISP, if you have one.
>   FURTHERMORE, shim6 will let you failover active connections from one
> host to another without starting a new TCP connection.

There have been more than a few talks/panels/discussions of shim6/ipv6
at previous nanog, arin, IETF meetings. The situation is far from this
simple for production enterprise networks. It's even less simple for
ISP-type networks.

SHIM6 seems to fit certain niches, I'm not sure that enterprise
multi-homing nor 'provider' multihoming are them. Content provider
multihoming seems even less likely to use shim6...

>
>   The problem is that, if you have a 1000 hosts, and they are on
> multiple subnets (perhaps at multiple locations via VPNs or leased
> lined, or ad-hoc wireless networks, or...), that you'd like to continue
> to be able to address them even when your ISP is dead, etc.
>

in today's world that's a NAT, VPN, multi-homed solution... tomorrow
that may not be the case, this is all to be determined. (and yes, NAT
isn't even required in the existing case, but... )