Re: A proposal for draft-ietf-6man-rfc4291bis-07

[David Farmer <[e-mail]>]

On Thu, Mar 2, 2017 at 6:18 PM, Manfredi, Albert E <
[e-mail]> wrote:

> From: ipv6 [mailto:[e-mail]] On Behalf Of David Farmer
>
> > I've been thinking about this, really hard, and listening to what
> > others are saying. I think we want to say the following;
> >
> > 1. IPv6 unicast routing is 128 bits in length [BCP198], AKA not classful!
>
> Okay.
>
> > 2. Subnet Prefixes of 64 bits are RECOMMENDED
>
> If a home or other network is assigned a /64, what do we recommend for
> internal subnetting, then? Can't be done? Or does the IETF forbid ISPs from
> assigning /64s?
>

First of all, a home network is essentially a site, and this is an issue
discussed in section 2.4.4 which probably needs to be updated too.  I think
including a direct reference to RFC 6177 would be a good idea, you get
there indirectly via the reference to RFC3587 and it's reference to RFC3177
which was updated by RFC6177. Furthermore, HOMENET RFC7368 says

   The introduction of IPv6 for home networking makes it possible for
   every home network to be delegated enough address space from its ISP
   to provision globally unique prefixes for each such subnet in the
   home.  While the number of addresses in a standard /64 IPv6 prefix is
   practically unlimited, the number of prefixes available for
   assignment to the home network is not.  As a result, the growth
   inhibitor for the home network shifts from the number of addresses to
   the number of prefixes offered by the provider; this topic is
   discussed in BCP 157 [RFC6177], which recommends that "end sites
   always be able to obtain a reasonable amount of address space for
   their actual and planned usage."

However, if a provider only delegates a /64, this new text ensures that
prefix could be further subnetted down below /64 using manual config or
possibly DHCPv6.  This is NOT RECOMMENDED, /64 subnets are RECOMMENDED and
RFC6177 clearly RECOMMENDS sites be allow more than one /64 subnet.


> > 3. IIDs are REQUIRED to be 64 bits
>
> I thought this interminable last call was specifically because many of us
> object to this idea. With those exceptions such as SLAAC on Ethernet. Or
> maybe IIDs should not be longer than 64 bits?
>

Variable length IIDs are a BIG CHANGE and it will break current code, I'm
not sure that is a really good idea.



> > 4. Say nothing that would imply you cannot configure a prefix length
> longer than 64 bits in length, at least manually.
>
> Agreed, but does that not invalidate #3?
>

Nope, the new text I supplied basically says 64 bit IIDs doesn't
necessarily imply every subnet on the Internet is a /64. All the way back
to RFC1884 this text says "at a minimum, a node may consider that unicast
addresses (including its own) have no internal structure", and an IID is
just that internal structure of an address, which is OPTIONAL by that text.


> Bert
>
>


-- 
===============================================
David Farmer               Email:[e-mail]
Networking & Telecommunication Services
Office of Information Technology
University of Minnesota
2218 University Ave SE        Phone: [phone]
Minneapolis, MN 55414-3029   Cell: 612-812-9952
===============================================