Re: [GLOBAL-V6] Comments on AP Consensus

[Gert Doering <gert@Space.Net>]

Hi,

some remark first - I am wondering why this discussion is revived now.
About all arguments have been discussed to death in January / February
already, and in the end, some sort of consensus seems to have been
achieved.

On Tue, Apr 16, 2002 at 09:59:24AM +0100, Dave Wilson wrote:
> > In addition, with IPv6, people *will* *not* accept end sites' routes
> > in the default free zone.  Some providers that already do IPv6 have
> > pretty clearly stated that.
> 
> Yes. This, combined with the proposal under discussion, combined with 
> the fact that the multihoming groups have not yet reported, means that 
> in effect the RIRs are now effectively deciding who may have the ability 
> to decide their own routing policy. This was not the case in IPv4.

This not much different from IPv4.  If you get a /24 PI in IPv4, and people
do not accept your /24 due to "why should I burden my routers with this",
you're not routed either.   In addition, PI space is already assigned today
without concern for routeability, only on a as-needed basis.

The difference in IPv6 is that there is no PI space, and thus, no
pretense of "here you have some IP space for yourself, but don't consider
it useful in any way".

PI is *not* useful.  It is harmful, and costly for us all.

> Any number of /48 assignments, be it 776 or 200 or any number greater 
> than 1, is a great mechanism for deciding who will make lots of 
> assignments (DSL make lots, NRENs make few) but is not a good mechanism 
> for deciding how many actual addresses will be used, and definitely not 
> a good policy for deciding who may multihome.

My proposal was different ("everybody who fulfills the necessary paperwork
to become a LIR and then signes that they will assign /48s to third parties",
no mention of numbers).  But this proposal was shot down by the Americans,
and the APNIC people settled for this compromise.

We need some sort of global policy, and we need it *now*.  

Immediately afterwards, discussion can start on policy revisal.

> Under this policy, a small or medium-size NREN would not be able to act 
> as a bridge between the high speed GEANTs or Abilenes and the general 
> internet, something that is crucial to our customers at least and would 
> definitely be a force against deployment. 

Why would that NREN need their own /35 for that, if they only serve
as transit provider for third party networks that already have their
own IPv6 space?  They could use a /48 from either of them.

> Unless that NREN happens to 
> connect 200 small schools which can be given a /48 each. This seems to 
> me to be very broken.

If a NREN connects a bigger university and that university connects 
student hostels, a fair number /48s can get used up pretty quickly.

But in general, I agree that this is not perfect (but 200 is a lot better
than 776).

[..]
> The consensus seems to be that HD is a good measurement system (I would 
> like to see the discussion on this because I don't agree that it's 
> measuring what people want it to measure). If we try to keep this then, 
> perhaps we could replace

I don't see anything useful in doing this.  All you can achieve *now* 
is "delay things for another four month".  If you change the proposal
document now, it has to wait for the next ARIN meeting to get consensus
- which is four month hence.

Besides, your proposal is just overly complex and doesn't help your
"IPv6 NREN with no IPv4 infrastructure and no wish to assign lots of
/48s to end sites" either - for a simple proposal for those cases, mine 
would have been fine, but people did not want that.  

We have been through all of this already...

Gert Doering
        -- NetMaster
-- 
Total number of prefixes smaller than registry allocations:   74810  (74196)

SpaceNet AG                 Mail: netmaster@Space.Net
Joseph-Dollinger-Bogen 14   Tel : +49-89-32356-0
80807 Muenchen              Fax : +49-89-32356-299

-
- This list (global-v6) is handled by majordomo@lists.apnic.net