<<< Chronological >>> Author Index    Subject Index <<< Threads >>>

Re: [hostmaster-staff] Re: MIR proposal

  • To: "Stephen Burley" < >
  • From: Sabrina Waschke < >
  • Date: Thu, 27 Sep 2001 19:11:25 +0200
  • Cc: "Anne Lord" < >
    "Hamid Alipour" < >
    "Mirjam Kuehne" < >
  • Reply-to:

 "Stephen Burley" stephenb@localhost writes:
 * A question to the NCC or any other registry managers:
 * 
 * What is the criteria by which the RIR's request space from IANA, is it an
 * 80% usage rule?

Yes, it is.

Regards,
Sabrina Waschke

-- 

o------------------------------------------o
| Sabrina Waschke         sabrina@localhost |
| Registration Services Operations Manager |
|                                          |
| RIPE NCC             tel +31 20 535 4444 |
| www.ripe.net         fax +31 20 535 4445 |
o------------------------------------------o


 * ----- Original Message -----
 * From: "Anne Lord" anne@localhost
 * To: "Hamid Alipour" alipour@localhost
 *; "Mirjam Kuehne" mir@localhost
 * Sent: Thursday, September 27, 2001 8:36 AM
 * Subject: Re: [hostmaster-staff] Re: MIR proposal
 * 
 * 
 * 
 * hi,
 * 
 * > Stephen seems just wants to solve UUNET problem
 * > with proposing MIR. However I am agree basically
 * > with the Idea. APNIC has added NIR
 * > ( National Internet Registry ) to the hierarchy.
 * > I think RIPE must let the NIRs as well.
 * 
 * Just a note about this. The membership category of NIR actually
 * does not relate in any way to the specific problem that Stephen is trying
 * to address which is that of large multinational organisations routed
 * under one AS, having discontiguous IP address allocations through the
 * establishment of many LIRs. In fact, the NIR model actually does nothing
 * for aggregation - as NIRs receive a block which they further allocate
 * to their members who run businesses within a particular country. The
 * members in those countries served by NIRs are more likely to receive
 * discontiguous blocks (simply because the NIRs have a smaller pool), thus
 * not contributing to aggregation of routing information at all.
 * We are working with the NIRs to solve this with a referral process for
 * allocations directly from APNIC for the very large members of NIRs.
 * 
 * Of course, having access to a local language service is very much on the
 * plus side of having NIRs.
 * 
 * For the record though, the NIRs exist under the confederation membership
 * category. This also includes ISP confederations as well as NIRs. The
 * two are *very* different entities, so the confederation category has
 * been suspended until we work out a better solution.
 * 
 * While I agree totally with Stephens objective and understand the motivation,
 * the proposal needs detail. How would it work exactly? APNIC's ISP
 * confederation
 * model which tried to address the same thing, did not work, and gave unfair
 * advantages to the ISP confederations. (Part of the reason the
 * 'confederation'
 * category has been suspended).
 * 
 * It is definately a laudable challenge to try to produce a model and
 * procedures such that the policies are fairly applied to all.
 * 
 * regards
 * 
 * Anne
 * _____________________________________________________________________
 * Anne Lord, Manager, Policy Liaison                  anne@localhost
 * Asia Pacific Network Information Centre       phone: +61 7 3367 0490
 * http://www.apnic.net                          fax:   +61 7 3367 0482
 * _____________________________________________________________________
 * 
 * 
 * 
 * 
 * >
 * >
 * > ----- Original Message -----
 * > From: "Stephen Burley" stephenb@localhost
 * > To: crain@localhost; lir-wg@localhost
 * > Sent: 06/09/2001 7:40 È.Ù
 * > Subject: Re: MIR proposal
 * >
 * >
 * > >
 * > > ----- Original Message -----
 * > > From: "John L Crain" crain@localhost
 * > >
 * > > Sent: Thursday, September 06, 2001 4:03 PM
 * > > Subject: Re: MIR proposal
 * > >
 * > >
 * > > > <CUT>
 * > > >
 * > > > Hi Gert,
 * > > >
 * > > > >
 * > > > > And yes, this is also very much needed for IPv6.  Getting a /35 and
 * > > > > having to hand out individual /48's to customers of customers of
 * ours
 * > > > > isn't going to build proper hierarchical routing.
 * > > >
 * > > > The concepts for IPv6 that are under discussion do already cover this.
 * > > > An allocation goes to a large ISP who can then assign /48's directly
 * to
 * > > > networks connecting to them or shorter prefixes to
 * > resellers/downstreams.
 * > > >
 * > > > I'm not sure if this works in IPv4 because of the limited amount of
 * room
 * > > we
 * > > > have to play with.
 * > >
 * > > We are only limited because of teh current thinking and structure.
 * > >
 * > >
 * > > >
 * > > > I'm also not sure what the criteria would be in the proposal that
 * > defines
 * > > > who is and isn't allowed to become a MIR. It's certainly a differnet
 * > > concept
 * > > > to the present one in the RIPE region where LIR's don't "officially"
 * > sub-
 * > > > allocate.
 * > > >
 * > >
 * > > Its not so different from the RIR model.
 * > >
 * > > > I can certainly see why a large ISP would want to do this. I'm not
 * sure
 * > > how
 * > > > it changes the dynamics for smaller ISP's as to how they would get
 * their
 * > > IP
 * > > > addresses. Becoming an LIR with an upstream rather than a regional
 * > > registry
 * > > > I assume means renumbering if you change the upstream.
 * > > >
 * > >
 * > > MIR's are only to be created within a network (AS if you like) they
 * would
 * > > not suballocate to customers only LIR's withing their network (usualy
 * > > country specific). Other LIR's not needing a MIR would deal direct with
 * > the
 * > > NCC. UUNET has 17 LIR's currently the MIR would suballocate to these not
 * > to
 * > > other ISP's or customers direct.
 * > > BTW Nice to hear from you.
 * > >
 * > >
 * > > > John Crain
 * > > >
 * > > >
 * > > >
 * > > >
 * > > >
 * > > >
 * >
 * > *    Mailing List: hostmaster-staff                                    *
 * > *    Handled by majordomo@localhost                              *
 * >
 * 
 * 
 * 







  • Post To The List:
<<< Chronological >>> Author    Subject <<< Threads >>>