[dhcwg] Re: [v6ops] Re: [IPv6]Re: Android now su pports DHCPv6 PD

>
> If but anyone had warned of this outcome before.
>
Ha, I've been doing /44s per-site as a minimum policy in all my IPv6
projects for years, and was sneezed at by some very members of this list a
while ago, IIRC. Yet here we are talking about larger prefixes.

Quick recap for those who may not be familiar with what I suggested as a
*minimum* (in the past):
/32 per org (purely for backbone-only, no customer
links/VMs/Hosts/End-users are ever part of this global /32) — in other
words, if any law enforcement comes knocking at your door, we should never
see a request originating from this /32 if we can help it. The prefix is
for the infra, not human use.
/44 per backbone PoP/Site, even your smaller colos near the street cabinets
for SPs.
/48 per parent aggregate function, then proceed to go smaller depending on
the function and sub-function, but nothing smaller than /64 excluding
shared /64 for loopbacks (/128 per loopback).

Separate /32s for customer/human-use subnetting, the number of /32s will
vary based on business scale, type, scope etc; CSP vs ISP vs carrier-only
SP vs LTE/5G carrier vs tiny WISP in the middle of nowhere vs niche cases
like a “Stock market Trading-network-only VPS provider, no EVPN/VXLAN, No
K8s, No containers, No Docker, just plain L3 unicast BGP to the host
hypervisor” (an interesting use-case that came to me recently) etc.

I'm pro-routing (L2 should be minimised in a design), pro-IPv6 and I think
ia_pd (as it stands today) is probably the closest (let's not get into a
debate on what defines a routing protocol though) thing to es-is, i.e.
route everything as much as possible, down to your endpoints (phones,
laptops etc). I wonder if Lorenzo envisioned this all long to push for
global /48s? I don't know, but either way, I'm a fan of /48s (or larger),
life's easier. NAT66 and ULAs be damned!

For LTE/5G, I perhaps may be able to push an LTE/5G provider to do /63s per
UE, at least (if their 3GPP/EPC stuff is compatible, I'm no expert on that
side of the fence). I'll be sure to share my findings from real-life
hands-on ops, should I come across anything interesting.

*--*
Best Regards
Daryll Swer
Website: daryllswer.com
<https://l.shortlink.es/l/288dacdda2ad7df49bdbe485f22b48d053ebd405?u=2153471>


On Tue, 16 Sept 2025 at 20:32, Gert Doering <[email protected]> wrote:

> Hi,
>
> On Tue, Sep 16, 2025 at 06:04:33PM +0900, Lorenzo Colitti wrote:
> > Recommendations for network operators are written in RFC 9663. Some text
> > about expected prefix lengths is in section 8 of that RFC. RFC 9762 says
> > the prefix must be SLAAC-sized, which currently means it must be a /64
> per
> > device. A /48 is fine for a small or medium network, but a campus with
> tens
> > of thousands of devices on it probably needs more than that.
>
> If but anyone had warned of this outcome before.
>
> Gert Doering
>         -- NetMaster
> --
> have you enabled IPv6 on something today...?
>
> SpaceNet AG                      Vorstand: Sebastian v. Bomhard,
>                                            Karin Schuler, Sebastian Cler
> Joseph-Dollinger-Bogen 14        Aufsichtsratsvors.: A. Grundner-Culemann
> D-80807 Muenchen                 HRB: 136055 (AG Muenchen)
> Tel: +49 (0)89/32356-444         USt-IdNr.: DE813185279
>
> _______________________________________________
> v6ops mailing list -- [email protected]
> To unsubscribe send an email to [email protected]
>

_______________________________________________
dhcwg mailing list -- [email protected]
To unsubscribe send an email to [email protected]