RE: Confirmation to advance: draft-ietf-6man-ipv6only-flag-05

["Manfredi (US), Albert E" <[e-mail]>]

From: ipv6 <[e-mail]> On Behalf Of David Farmer

> The two alternatives to this flag I've heard discussed are;
>
> 1. Layer 2 filtering of Etherthypes 0x0800 and 0x0806
> 2. RFC2563 to signal host to not use Link-local IPv4

And/or, do not return A records after DNS queries, switch off DHCPv4 entirely, do not make servers reachable even with statically configured IPv4 addresses, do not allow routers to respond to ARP queries. Let the hosts decide for themselves, how long and hard they want to keep trying to use IPv4. Anyone worried about battery power can certainly take those hints! Those not worried, no problem.

The intention of this flag, from those who thought it was important, has always been to prevent *all IPv4 traffic*. And yet, that remains unachievable anyway. You, the user, even if on a well-managed enterprise network, in your office or in your lab, can always install your own IPv4-only peripheral device, to reach with your own dual-stack host. So flag or no flag, you won't necessarily achieve this strict IPv6-only goal.

And quite frankly, who cares? In most networks, managed or unmanaged, no one should care. If you manage an enterprise network, the best you can do is only set up enterprise hosts, servers, and routers, for only IPv6. Employ the measures listed above, and you're as close to being done as you'll ever be. And no new flag needed.

Hey, whether you create a new flag or not, you will require updated logic in hosts, to save battery power. Let the hosts which care about this update themselves, without needing any new flag, and leave the others be.

Bert