[arin-ppml] Change of Use and ARIN (was: Re: AFRINIC And The Stability Of The Internet Number Registry System)
owen at delong.com
Mon Sep 13 17:06:32 EDT 2021
> On Sep 13, 2021, at 12:57 , Fernando Frediani <fhfrediani at gmail.com> wrote:
> Hi Bill
> How would it be possible that something marked as *Future* Use be blocked at the firmware level on any network equipment ?
> If it says Future it means one day it would be used for something. It could be for multicast it could be for Unicast, who knows. But adding a hard-coded to block any packets within this IP space simply thrown all in the bin.
> I would undestand if one treated that as bogons for some time, but not by having the idea to embed that in the firmware.
Because the future use is an undefined protocol capability that doesn’t yet exist. Obviously, one would not expect the addresses to be usable without a firmware update to the router that deals with the routing semantics involved in whatever new capability that is. Think how bad it would have been if when multicast started being tested out, every router on the planet by default forwarded such packets using unicast routing semantics? Instead, this previously future-use class D address space (224/3) was dropped-by-default and you couldn’t accidentally configure your router to treat it as unicast.
That turned out to be a pretty good safety feature, all things considered.
Today, we don’t foresee any development leading to a viable use for the remaining crumb of future-use in 240.0.0.0/4, but it’s not appropriate to say “we should have always just treated everything that we don’t know what we’ll do with as further GUA in software until we do.” That’s utterly nonsensical and very dangerous.
> Agree that IETF could have been less narrow and spent more time trying to find a usage that could have been beneficial to the Internet instead of just bet on the fast deployment of IPv6.
Meh… Don’t really see the point.
> Em 13/09/2021 15:32, William Herrin escreveu:
>> On Mon, Sep 13, 2021 at 10:54 AM Fernando Frediani <fhfrediani at gmail.com> wrote:
>>> I don't know who was the "genius" back in the past on network vendors
>>> who embedded to not forward traffic for that amount of /8's market as
>>> Future Use. I think that was one of the most disastrous decisions ever
>>> made in this area of IP space.
>> Hi Fernando,
>> IIRC, this was an IETF requirement. 240/4 is not reserved for future
>> unicast use, it's just plain reserved. The IETF could make it
>> multicast or some other odd thing tomorrow and any equipment treating
>> it as unicast would be malfunctioning.
>> Some systems treat it as unicast anyway. The Linux devs break the
>> standards in a number of adventurous ways. Not everyone does.
>> At any rate, blame for 240/4 not being usable for unicast belongs with
>> the IETF participants who have opposed designating it so, not with the
>> vendors who have complied with the IETF's designation.
>> Bill Herrin
> You are receiving this message because you are subscribed to
> the ARIN Public Policy Mailing List (ARIN-PPML at arin.net).
> Unsubscribe or manage your mailing list subscription at:
> Please contact info at arin.net if you experience any issues.
More information about the ARIN-PPML