[IPv6]Re: Analysis of Ungleich ULA Registry

[Bob Hinden <[e-mail]>]

Mark,

> On May 22, 2024, at 8:55 PM, Mark Smith <[e-mail]> wrote:
>>> 
>> 
>> 
>> Indeed, we are not. Nevertheless, if the data is to be believed, people are, in fact, using multiple /48s. I'm simply saying that is a fact, and it counters the argument that "no one needs more than a /48," at least in my opinion.
>> 
> 
> I think you can only conclude that if you believe that most people are
> generating ULA /48s correctly. I've seen enough examples that they

I note that it’s probably not “people” doing this, it is routers.  On my home network, the routers (eero) creates what looks to me to be a ULA /48 with a random prefix.

> aren't such that I don't assume anymore that individual people are
> following RFC 4193 correctly unless I see some signs or evidence that
> they are.
> 
> The two main reasons to need a larger prefix that a current one are:
> 
> - you don't have enough address space / subnets
> 
> - you need to aggregate at points within your network to scale either
> your routing protocol or your router's FIBs.
> 
> With a single /48 ULA providing 65 536 /64s, with many organizations
> not having many 10s of 1000s of links to number within their network,
> with BGP easily scaling to 65 536 routes without aggregation, with a
> network using BGP for large scale routing likely to have routers with
> large FIB capacities, with many organisations not having 10s of 1000s
> of hosts to give individual /64s if that was the deployment choice,
> and with other scaling techniques like using an EVPN with ND/ARP proxy
> (RFC 9161) to put many hosts inside a single /64 subnet, I don't think
> shorter than /48 ULA is actually a common requirement at all.

I agree with that.

I would go further and say that organizations that require shorter prefixes, should probably just get Global addresses from their ISPs or from the RIRs.

Bob