- address-policy-wg - mailman.ripe.net

RE: [address-policy-wg] RE: Private address space in IPv4 and IPv6 [was something irrelevantly titled]
by poty＠iiat.ru 01 Jun '09

01 Jun '09

> -----Original Message----- > From: address-policy-wg-admin(a)ripe.net [mailto:address-policy-wg- > admin(a)ripe.net] On Behalf Of Gert Doering > Sent: Saturday, May 30, 2009 4:00 PM > To: michael.dillon(a)bt.com > Cc: address-policy-wg(a)ripe.net > Subject: Re: [address-policy-wg] RE: Private address space in IPv4 and > IPv6 [was something irrelevantly titled] > > Hi, > > On Fri, May 29, 2009 at 11:03:16AM +0100, michael.dillon(a)bt.com wrote: > > A very large number of organizations depend on these > > internetworks, and they would not be terribly happy if > > ISPs would hijack the entire IP address space for their > > own profits. But I think that the RIR boards understand > > this and have no intention of changing the rules to > > reserve IP addresses only for the public Internet. > > Let me second that. > > This has always been my understanding on the principles that govern > the RIRs' operation - "provide unique numbers to the people that need > unique numbers". Be it IPv4, IPv6 or ASes. No, because it has (providing the "numbers") the scope - the Internet, If I want to create my own network, not connected to the scope, I don't need to use any of rules of disconnected network. I'm surprised that you are not understanding and differentiating that. Maybe, following your thoughts, I could get the IP number for my pet? Just to be more "connected" to the network? > > Gert Doering > -- APWG chair > -- > Total number of prefixes smaller than registry allocations: 128645 > > SpaceNet AG Vorstand: Sebastian v. Bomhard > Joseph-Dollinger-Bogen 14 Aufsichtsratsvors.: A. Grundner- > Culemann > D-80807 Muenchen HRB: 136055 (AG Muenchen) > Tel: +49 (89) 32356-444 USt-IdNr.: DE813185279 > Vladislav Potapov Ru.iiat

2 1

RE: [address-policy-wg] RE: Private address space in IPv4 and IPv6 [was something irrelevantly titled]
by poty＠iiat.ru 01 Jun '09

01 Jun '09

Yes, Wilfried, interesting, but it seems that YOU are behind the reality, not mine. > -----Original Message----- > From: Wilfried Woeber, UniVie/ACOnet [mailto:Woeber@CC.UniVie.ac.at] > Sent: Saturday, May 30, 2009 12:51 AM > To: Potapov Vladislav > Cc: address-policy-wg(a)ripe.net > Subject: Re: [address-policy-wg] RE: Private address space in IPv4 and > IPv6 [was something irrelevantly titled] > > poty(a)iiat.ru wrote: > > > IPv4 exists most of it's time without RIR at all. So it's not prove > > anything. > > May I suggest that you read up on reality, in particular the period > of substantial growth and wide-spread penetration? > > > Other consideration is only philosophy, not policy. > > Interesting... > My feeling is that we should go back to real-life issues. > > Wilfried.

1 0

Re: [address-policy-wg] RE: Private address space in IPv4 and IPv6 [was something irrelevantly titled]
by Andy Davidson 31 May '09

31 May '09

On 31 May 2009, at 11:28, Olivier MJ Crepin-Leblond wrote: > "Andy Davidson" <andy(a)nosignal.org> wrote: > >> IP networks can't function properly (or at least, as intended) when >> consumers of addresses start using addresses that are already >> assigned to others already. >> >> We shall see this be a problem for networks who ignore this warning >> as the last v4 /8s start to be assigned. > > Actually who is mandated to act as a policeman when this happens? > Is it IANA? Is it the RIRs? Is it informal routing agreements > outside the realm of these organizations? Nobody - just as nobody can guarantee reach-ability. Andy

1 0

RE: [address-policy-wg] RE: Private address space in IPv4 and IPv6 [was something irrelevantly titled]
by poty＠iiat.ru 31 May '09

31 May '09

See in-line: > -----Original Message----- > From: address-policy-wg-admin(a)ripe.net [mailto:address-policy-wg- > admin(a)ripe.net] On Behalf Of michael.dillon(a)bt.com > Sent: Friday, May 29, 2009 2:03 PM > To: address-policy-wg(a)ripe.net > Let me describe a real situation that we have with some > of our customers. One of several IP networks that we > operate is called Radianz. This network has PoPs in > 120 countries and has over 3000 customers connected > to it, many of them with multiple connections in different > cities. This network is an internetwork because it > interconnects the networks of 3000 other organizations. > In order to function, this network requires globally > unique IP addresses to be used for each of its > member networks even though it is not connected to > the Internet. > > The Radianz network is not a private network. It is > also not a public network. But it is a SHARED network > just like the public Internet. Then Radianz could easily create its own rules without bothering the World, couldn't it? And so - use ANY IP addresses. Why should I see the internal networks (I use corrected "private" meanings) of Radianz or other such companys? If it is NEVER interact with my or the most of other networks in the Internet? > Historically, IANA has > always allocated globally unique IP addresses to > organizations on this type of shared network. When > RFC 2050 was written by authors from the RIRs, including > RIPE, they included this text: > > the organization has no intention of connecting to > the Internet-either now or in the future-but it still > requires a globally unique IP address. The organization > should consider using reserved addresses from RFC1918. > If it is determined this is not possible, they can be > issued unique (if not Internet routable) IP addresses. > Now, the "historically" invented state should be corrected. Not all IPv4 practice should be automatically go to IPv6. Several of them are too company-specific rather than serve ALL community. > > A very large number of organizations depend on these > internetworks, and they would not be terribly happy if > ISPs would hijack the entire IP address space for their > own profits. But I think that the RIR boards understand > this and have no intention of changing the rules to > reserve IP addresses only for the public Internet. A very large, but not all! A very large number of organization (I think much more than for Radianz) depends on the services, not presented here and not gain any good from that. Vladislav Potapov Ru.iiat

2 1

2009-07 New Policy Proposal (Internet Assigned Numbers Authority (IANA) Policy for Allocation of ASN Blocks (ASNs) to Regional Internet Registries)
by Filiz Yilmaz 31 May '09

31 May '09

PDP Number: 2009-07 Internet Assigned Numbers Authority (IANA) Policy for Allocation of ASN Blocks (ASNs) to Regional Internet Registries Dear Colleagues A new RIPE Policy Proposal has been made and is now available for discussion. You can find the full proposal at: http://www.ripe.net/ripe/policies/proposals/2009-07.html We encourage you to review this proposal and send your comments to <address-policy-wg(a)ripe.net> before 24 June 2009. Regards Filiz Yilmaz Policy Development Manager RIPE NCC

5 4

RE: [address-policy-wg] RE: Private address space in IPv4 and IPv6 [was something irrelevantly titled]
by poty＠iiat.ru 30 May '09

30 May '09

I think you intentionally make confusion for everyone. ----Your---------- This is wrong. The Internet Protocol (IP) was not invented only to create the public Internet. It was created so that all networks, that want to interconnect with another network, have an internetworking protocol that does the job. The public Internet is the biggest internetwork but it is not the only one. Globally unique addresses are required so that a network can interconnect with other networks without renumbering. --------------- Nobody, including me, prevents you or someone else to using IP addresses! The term "Globally unique addresses" is useful if you mean PUBLIC part of the network. In my own network I can use the address I like, not involving the RIPE or any other organizations and rules. If I want to interconnect with another network I should agree with the network only, not with all the world. ----Your---------- The RIRs are not there to serve the Internet. They are there to serve the users of IP technology which is a bigger group than just the Internet. It is OK for RIPE to provide services that are only needed on the Internet, but it is not nice if RIPE would stop providing services to IP users who are not connected to the Internet. --------------- Not at all: "The RIPE NCC is an independent, not-for-profit membership organisation that supports the infrastructure of the Internet through technical co-ordination in its service region. The most prominent activity of the RIPE NCC is to act as the Regional Internet Registry (RIR) providing global Internet resources and related services (IPv4, IPv6 and AS Number resources) to members in the RIPE NCC service region." The users, that not connected to the Internet, are not in scope of the RIPE. IP is not the only network technology. And the network is not the only activity in the World.

3 2

RE: [address-policy-wg] RE: Private address space in IPv4 and IPv6 [was something irrelevantly titled]
by poty＠iiat.ru 29 May '09

29 May '09

IPv4 exists most of it's time without RIR at all. So it's not prove anything. Other consideration is only philosophy, not policy. > -----Original Message----- > From: address-policy-wg-admin(a)ripe.net [mailto:address-policy-wg- > admin(a)ripe.net] On Behalf Of michael.dillon(a)bt.com > Sent: Friday, May 29, 2009 2:29 PM > To: address-policy-wg(a)ripe.net > Subject: RE: [address-policy-wg] RE: Private address space in IPv4 and > IPv6 [was something irrelevantly titled] > ... > > IPv4 runout proves that the historical ways are the right ways. > > --Michael Dillon >

3 2

RE: [address-policy-wg] RE: Private address space in IPv4 and IPv6 [was something irrelevantly titled]
by poty＠iiat.ru 29 May '09

29 May '09

Not each other, but all to some! From: David Freedman [mailto:david.freedman@uk.clara.net] Sent: Friday, May 29, 2009 2:06 PM To: Potapov Vladislav; jeroen(a)unfix.org Cc: nick(a)inex.ie; frederic(a)placenet.org; address-policy-wg(a)ripe.net Subject: RE: [address-policy-wg] RE: Private address space in IPv4 and IPv6 [was something irrelevantly titled] >It's the problem of the merging banks or any other companies, not the >whole World. Why RIPE should do the work for them? I distill that argument down to "why should we help eachother", this is starting to go seriously off-topic now. Dave.

2 1

RE: Private address space in IPv4 and IPv6 [was something irrelevantly titled]
by poty＠iiat.ru 29 May '09

29 May '09

Jeroen, The clash is about: * RIR-space + guaranteed globally unique + CAN be routed on the internet - you will have to do paperwork and pay for it My point of view: RIR-s space is for routing on the Internet. Not for private use! So it MUST be routed on the Internet. And private networks should invent their own rules, personally I will not object that as far as it is not affect my access to public part of the Internet! -----Original Message----- From: Jeroen Massar [mailto:jeroen@unfix.org] Sent: Friday, May 29, 2009 12:37 PM To: Potapov Vladislav Cc: david.freedman(a)uk.clara.net; nick(a)inex.ie; frederic(a)placenet.org; address-policy-wg(a)ripe.net Subject: Private address space in IPv4 and IPv6 [was something irrelevantly titled] poty(a)iiat.ru wrote: > Maybe you are right, but it doesn't prove that is IS good in IPv6 > world too. I can't understand, why I should think about such private > matters (and indirectly fund this) and count it as arguments in the > RIPE's policy development? If the allocation will never be announced > to the public network called the Internet, then it's not the scope of our thinking! IPv4: * RFC1918 + just grab - everybody in the world uses it, lots of clashes - not suitable for interconnecting ever to other networks - generally implies a lot of NAT at one point in time * RIR-space + guaranteed globally unique - you will have to do paperwork and pay for it IPv6: * ULA, RFC4193 + nobody to talk to, calculate your own - never to be used anywhere on the Internet - not 100.00000% sure that it is globally unique (also see http://www.sixxs.net/tools/grh/ula/ for a 'registry' which would make it at least "unique" when everybody uses that) - could imply NAT, though that should not be used with IPv6 * RIR-space + guaranteed globally unique + can be routed on the internet - you will have to do paperwork and pay for it You can pick what you want, but heed the warnings. Greets, Jeroen

4 3

RE: [address-policy-wg] RE: Private address space in IPv4 and IPv6 [was something irrelevantly titled]
by poty＠iiat.ru 29 May '09

29 May '09

See in-line-2 > -----Original Message----- > From: Jeroen Massar [mailto:jeroen@unfix.org] > Sent: Friday, May 29, 2009 12:57 PM ... > You are going to require that people route everything onto the > Internet? > Not going to happen. There are a lot of assigned blocks which you will > never ever see on the Internet. And why would they be, it is their > network, thus theirs to route or not, to firewall or not. > 1. "assigned" blocks is not the things, that involves the routing. If you mean allocations - say that. 2. If it is THEIR networks, why the need GLOBALLY unique addresses? Then they could use ANY addresses they like not asking me, or you, or RIPE! > > And private networks > > should invent their own rules, personally I will not object that as > > far as it is not affect my access to public part of the Internet! > > You mean a separate registry so that when people are "OH I WANT > INTERNETZ" that they simply announce their prefix, which clashes with > real prefixes on the Internet !? That will be a lot of fun. I think you can't understand what is difference between private usage and announcing the block in the Internet. If you want to use some IP addresses privately, you can do it without any RIR. And only in the case you want to announce some block to the public network, you have to abide the Internet rules and get unique numbers! > The sole reason for having registries in the first place is to make > sure these little numbers are globally unique and that they thus don't > clash. > > Ever tried to merge the network of a couple of banks after they where > acquired by each other and all where using 192.168.0.0/16 in their > internal "totally private" networks? > > Uniqueness is what is needed there for that to work. > It's the problem of the merging banks or any other companies, not the whole World. Why RIPE should do the work for them? Or take in the consideration the internal problems of SOME cases into the whole region policy? Vladislav Potapov Ru.iiat

2 1