[lir-wg] Re: [ipv6-wg at ripe.net] IPv6 TLAs for mobile operators

At 05:39 PM 14-11-02 +0100, Sabrina Wilmot wrote:

5.1.1 is a problem for me.  To quote:
a) be an LIR;
b) not be an end site;
c) plan to provide IPv6 connectivity to organisations to which it will 
assign /48s, by advertising that connectivity through its single aggregated 
address allocation; and
d) have a plan for making at least 200 /48 assignments to other 
organisations within two years.

I am an NREN comprising 8 universities.  Not an ISP.  How are we supposed 
to get a /48 or a /32 for research purposes that can be used for multihoming?

Thanks,
Hank


>Mobile operators participated in IPv6 policy development to ensure
>requirements were met. As was pointed out the RIPE NCC has made an IPv6
>allocation to a mobile-service provider, and they were able to meet the
>current policy's criteria.
>
>If others feel the policy needs clarification, we encourage the community
>to continue discussing this issue of the current policy and how it would
>apply to a mobile-only operator using Gert Doering's  proposal as a
>starting point.
>
>I suggest to limit this discussion to the LIR-WG list.
>
>Kind regards,
>
>Sabrina Wilmot
>--
>
>o------------------------------------------o
>| Sabrina Wilmot          sabrina at ripe.net |
>| Registration Services Operations Manager |
>|                                          |
>| RIPE NCC             tel +31 20 535 4444 |
>| www.ripe.net         fax +31 20 535 4445 |
>o------------------------------------------o
>
>
>On Mon, 4 Nov 2002 12:12:46 +0100
>Gert Doering <gert at space.net> wrote:
>
> > Hi,
> >
> > On Thu, Oct 31, 2002 at 06:57:05PM +0100, leo vegoda wrote:
> > [..]
> > > > But I really don't want to concentrate on that discussion. I want to
> > > > know if RIPE NCC accepts IPv6 TLA requests from mobile operators only
> > > > having GPRS/UMTS/WLAN customers - and therefore not providing
> > > > connectivity to organisations with a /48.
> > > > -> if anybody can clarify this, I would be very happy!
> > >
> > > The RIPE NCC implements the RIPE community's policy as described in
> > > the IPv6 Address Allocation and Assignment Policy. The policy was
> > > agreed at the RIPE 42 meeting in May.
> > >
> > > An LIR submitting a request for an initial IPv6 allocation will need
> > > to meet all four requirements specified in the policy.
> > >
> > >    <http://www.ripe.net/ripe/docs/ipv6policy.html#initial_criteria>
> >
> > This comment is not too helpful.  Of course those are the rules :-)
> >
> > The problem I see is that the policy explicitely specifies "assigns /48s",
> > which is what "we" assumed to be the generic case for the ISP -> customer
> > relation.
> >
> > On the other hand, RIPE-246 explicitly specifies the use of smaller
> > prefixes for special cases:
> >
> > --------------- quote ----------------
> > 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.
> >
> >    RIRs/NIRs are not concerned about which address size an LIR/ISP
> >    actually assigns. Accordingly, RIRs/NIRs will not request the
> >    detailed information on IPv6 user networks as they did in IPv4,
> > --------------- quote ----------------
> >
> > Looking at the number of potential IPv6 customers of a mobile network
> > operator, assigning each of them a /48 does't make much sense.
> >
> > On the other hand, even when assigning each end customer a /64,
> > and aggregating at cell boundaries (for example), the mobile network
> > is likely to make better usage of the IPv6 space than many smaller
> > ISPs that do match the letter of the "200 /48" rule.
> >
> > I, personally, think that a mobile network operator really should be
> > able to get an IPv6 allocation - if not them, who else?  And I also
> > see that at least one of them already has one (DE-D2VODAFONE,
> > 2001:0928::/32).
> >
> > So I think a clarification is needed - maybe the wording of the policy
> > document has to be changed to make very explicit that this is acceptable
> > usage, something like this:
> >
> > -------------- proposal ------------
> >      c) plan to provide IPv6 connectivity to organisations to which it
> >      will assign /48s or /64s according to 5.4.1, by advertising that
> >      connectivity through its single aggregated address allocation; and
> >
> >      d) have a plan for making at least 200 /48 assignments to other
> >      organisations within two years, or plan to assign "enough" (to be
> >      defined) /64s to be equivalent to 200 /48s.
> > -------------- proposal ------------
> >
> > one could apply HD ratio to the /64s inside the /48s, like "it's valid
> > if <hd-ratio>% /64s out of a /40 are assigned", a /40 being "about 200 
> /48s".
> >
> > Gert Doering
> >         -- NetMaster
> > --
> > Total number of prefixes smaller than registry allocations:  48540  (48282)
> >
> > SpaceNet AG                 Mail: netmaster at Space.Net
> > Joseph-Dollinger-Bogen 14   Tel : +49-89-32356-0
> > 80807 Muenchen              Fax : +49-89-32356-299
> >