nanog mailing list archives

Re: Q: is RFC3531 still applicable?


From: Mel Beckman <mel () beckman org>
Date: Thu, 16 May 2024 05:09:55 +0000

Jay,

Each IPv6 /64 should be thought of as the same as an IPv4 /32? That seems a tad wasteful. A single /64 has billions of 
times more addresses than the entire IPv4 address space. It is enough for any conceivable subnet. There are also 
billions of /56 prefixes available, so no ISP customer would ever exhaust those either. A customer can get as many /56s 
as they need.

The RFC seems to be concerned with aggregation efficiency, and while that may be a concern someday, so far computer and 
memory capacity has far outstripped prefix growth in the default-free zone.

If you can explain why a /64 would ever not be enough for a single subnet, I’m willing to listen.

 -mel

On May 15, 2024, at 9:52 PM, Jay Acuna <mysidia () gmail com> wrote:

A /64 is not "enough" period.  Each IPv6 /64 should be thought of as
the same as an IPv4 /32.
The RFC is still relevant.  You are able to be allocated IPs
justifying 8-bits per customer
(/56) and customers should expect that /56 be the minimum delegated by
their providers.

The prefix delegation for IPv6 is based on number of separate /64
subnets they might have a reason
to use (which can be for many reasons including security and division
of traffic and use cases),

Not number of individual hosts they may have, since subnet divisions
more granular than
/64 are not possible.

On Wed, May 15, 2024 at 8:17 AM Mel Beckman <mel () beckman org> wrote:
I never could understand the motivation behind RFC3531. Just assign /64s. A single /64 subnet has 
18,446,744,073,709,551,616  host addresses.  It is enough. Period.
-mel

--
-J

Current thread: