nanog mailing list archives
Re: 240/4 (Re: 44/8)
From: George Herbert <george.herbert () gmail com>
Date: Mon, 22 Jul 2019 21:02:07 -0700
Most importantly, if you're running out of 1918 space is a totally different problem than running out of global routable space. If you patch common OSes for 240/4 usability but a significant fraction of say unpatched OSes, IOT, consumer routers, old random net cruft necessary for infrastructure aren't patched... it's not actually globally routable. At some point you can write off the few stragglers but... really, get IPv6 everywhere. On Mon, Jul 22, 2019 at 8:50 PM Owen DeLong <owen () delong com> wrote:
On Jul 22, 2019, at 20:14 , Mikael Abrahamsson <swmike () swm pp se> wrote: On Mon, 22 Jul 2019, Owen DeLong wrote:2. It was decided that the effort to modify each and every IPstack in order to facilitate use of this relatively small block (16 /8s being evaluated against a globalrun rate at the time of roughly 2.5 /8s per month, mostlyto RIPE and APNIC) vs. putting that same effort into modifying each and every IP stack to supportIPv6 was an equation of very small benefit for slightlysmaller cost. (Less than 8 additional months of IPv4 free pool vs. hopefully making IPv6 deployablebefore IPv4 ran out).Well, people are working on making 240/4 usable in IP stacks:https://raw.githubusercontent.com/dtaht/unicast-extensions/master/rfcs/draft-gilmore-taht-v4uniext.txtThere have been patches accepted into some BSDs and into Linuxtools/kernel and other operating systems to make 240/4 configurable and working as unicast space.I don't expect it to show up in DFZ anytime soon, but some people havedilligently been working on removing any obstacles to using 240/4 in most common operating systems.For controlled environments, it's probably deployable today with somecaveats. I think it'd be fine as a compliment to RFC1918 space for some internal networks.-- Mikael Abrahamsson email: swmike () swm pp seI guess people can do whatever they want. I personally consider it to be a sad sad waste of time that could be better spent deploying IPv6 to more places. Owen
-- -george william herbert george.herbert () gmail com
Current thread:
- Re: 44/8, (continued)
- RE: 44/8 Naslund, Steve (Jul 22)
- Re: 44/8 Stephen Satchell (Jul 22)
- Re: 44/8 Owen DeLong (Jul 22)
- 240/4 (Re: 44/8) Mikael Abrahamsson (Jul 22)
- Re: 240/4 (Re: 44/8) Owen DeLong (Jul 22)
- Re: 240/4 (Re: 44/8) George Herbert (Jul 22)
- Re: 240/4 (Re: 44/8) Ross Tajvar (Jul 22)
- Re: 240/4 (Re: 44/8) Greg Skinner via NANOG (Jul 26)
- Feasibility of using Class E space for public unicast (was re: 44/8) Doug Barton (Jul 26)
- Re: Feasibility of using Class E space for public unicast (was re: 44/8) William Herrin (Jul 26)
- Re: Feasibility of using Class E space for public unicast (was re: 44/8) Doug Barton (Jul 26)
- Re: Feasibility of using Class E space for public unicast (was re: 44/8) William Herrin (Jul 26)
- Re: Feasibility of using Class E space for public unicast (was re: 44/8) Doug Barton (Jul 27)
- Feasibility of using Class E space for public unicast (was re: 44/8) bzs (Jul 27)
- Re: Feasibility of using Class E space for public unicast (was re: 44/8) johnl (Jul 27)
- Re: Feasibility of using Class E space for public unicast (was re: 44/8) Randy Bush (Jul 27)
