[ppml] IPv6>>32

[Howard, W. Lee]

> -----Original Message-----
> From: owner-ppml at arin.net [mailto:owner-ppml at arin.net] On 
> Behalf Of Stephen Sprunk
> Sent: Monday, May 16, 2005 11:38 AM
> To: Owen DeLong; Larry J. Blunk; Michael.Dillon at radianz.com
> Cc: ppml at arin.net
> Subject: Re: [ppml] IPv6>>32
> 
> 

> "6.5.4.1   Assignment address space size
> Assignments are to be made in accordance with the existing 
> guidelines (RFC3177,RIRs-on-48), which are summarized here as:
> - /48 in the general case, except for very large subscribers
> - /64 when it is known that one and only one subnet is needed 
> by design
> - /128 when it is absolutely known that one and only one 
> device is connecting."
> 
> This appears to prevent ISPs (i.e. universities) from 
> assigning anything longer than a /48 to customers (i.e. 
> students) in most cases unless they do not qualify as "end sites".

Sounds to me like it's a design question.  Is dorm connectivity
designed to support multiple subnets per room?  If not (as I tend
to think) each port gets a /64.  I would expect dorms with 
multiple occupants to have multiple ports, regardless of whether
the switch is in the room or somewehre else, but if the university
prefers to install a switch with VLANs then a /48 is required.


> AFAICT, each student at a university qualifies as an "end 
> site".  Therefore, unless the specific criteria for a /64 or 
> /128 are met for each student, which is certainly not 
> guaranteed nor feasible to determine, each student must be 
> assigned a /48.
> 
> Interestingly, the above could also be (mis)read to say that 
> businesses need to assign a /48 to each employee.

Depends on the use of the network.  As the Manager of Network
Engineering for Stanley Associates, I would not assign a /48 to
each employee, because I don't want them installing alien crud
on my network.  YMMV

 
> > > Not to mention what actually deploying such a setup would 
> do to the 
> > > existing IPv4 addressing plan.  What are the odds ARIN 
> would approve 
> > > Merit (today, ignoring their legacy /8) requesting an IPv4 /12 so 
> > > they could give every student (38,000 cited elsewhere in 
> the thread) 
> > > a /29?  That's what we'd be forcing them to do if they had to 
> > > provide a subnet per student.
> >
> > This is where IPv4 != IPv6 and there does start to be some 
> reason to 
> > consider that addresses are no longer scarce.  In IPv4, it 
> is hard to 
> > imagine assigning a static address to every person on the 
> planet. In 
> > IPv6, it is hard to imagine the population being sufficient 
> that each 
> > person could not have a /64.
> 
> I'm working with the assumption that v6 and v4 will be 
> deployed with the same topology, which has yet to be 
> disproven.  If we give each student an IPv6 /64 (or /48), 
> then we'd need to give them an IPv4 /29 or shorter too.

I don't understand this assumption.  When I explain IPv6 to
people around me, I compare IPv6 /64 to IPv4 /32+NAT, but 
without the (arguably evil) complexities of NAT.  Clearly
the topologies can't be the same, because there's not enough
space in IPv4 to make it so.

More likely, IMHO, is for organizations using IPv4 with 10/8
NAT to use native IPv6 addresses under dual stacks.
 
> > > Perhaps some vendor will come out with a whiz-bang device that 
> > > allows a shared IPv4 subnet while routing IPv6 natively, 
> but I'm not 
> > > aware of anything like that on the market or even proposed for 
> > > development.
> >
> > I guess I'm not understanding the situation you are 
> describing here. 
> > There are several ways to do 6:4 gateways which would allow 
> you to run 
> > v4 and v6 on the same segment.
> 
> We're talking about native v4 and native v6, not some 
> gatewayed or NATed service.

I didn't think native IPv4 was specified; removing that part
of the conversation doesn't change the context any.

> > There is nothing which requires you to match your v4 
> topology to your v6 topology.
> 
> I'm not aware of any off-the-shelf consumer-grade solution 
> which allows them to be different.

There's no reason to match them, so there's no solution for
doing so.  At the organization's (LIR's/ISP's/university's)
edge they'll run native IPv6 and NATed IPv4; you wouldn't
expect that to be consumer grade.  Even if they're running
native IPv4, you can extend the depth of the network with
IPv6 without offering the same depth with IPv4 (i.e., you
can have subnets all the way down to /64).  Even if you want
matching topologies, IPv4 allows VLSM, so you'd only assign
a /29 (or /30) where it's needed, not necessarily to every
student.


> Keep in mind that most dorms (and hotels) currently allow 
> users to plug in a single PC and start working without any 
> sort of L3 device in the unit at all.  Any IPv6 model that 
> requires putting an L3 device in the unit is already making 
> things more expensive, possibly to the point of making the 
> transition not financially viable, and there doesn't appear 
> to be any way around that if we require a /48 per user or per 
> room.  Even a /64 per user or per room significantly raises 
> the administrative costs for both v4 and v6, but at least 
> that's technically feasible today.

There's still a layer 3 device, it's the NIC.  Maybe you meant
L3 router; it exists, it's just not in the dorm room.
Routing a /64 or /48 to a specific port does not mean there
has to be a router at the other end; it's perfectly reasonable
for a single host to be on that subnet.  

Lee