April 2008 - address-policy-wg

2007-08 New Draft Document Published (Enabling Methods for Reallocation of IPv4 Resources)
by Filiz Yilmaz 27 May '08

27 May '08

PDP Number: 2007-08 Enabling Methods for Reallocation of IPv4 Resources Dear Colleagues, The draft document for the proposal described in 2007-08 has been published. The impact analysis that was conducted for this proposal has also been published. This proposal outlines a framework to migrate previously allocated IPv4 resources from one Local Internet Registry (LIR) to another LIR within the RIPE NCC Service Region. You can find the full proposal at: http://www.ripe.net/ripe/policies/proposals/2007-08.html and the draft document at: http://www.ripe.net/ripe/draft-documents/ripe-424-draft2007-08.html We encourage you to read the draft document text and send any comments to address-policy-wg(a)ripe.net before 15 April 2008. Regards Filiz Yilmaz RIPE NCC Policy Development Officer

7 16

2008-02 Review Period extended until 16 May 2008 (Assigning IPv6 PA to Every LIR)
by Filiz Yilmaz 15 May '08

15 May '08

PDP Number: 2008-02 Assigning IPv6 PA to Every LIR Dear Colleagues, The Review Period for the proposal 2008-02 has been extended until 16 May 2008. With the acceptance of this proposal RIPE NCC will run a one-time operation to allocate an IPv6 block to every LIR that does not have any existing IPv6 holdings. You can find the full proposal at: http://www.ripe.net/ripe/policies/proposals/2008-02.html We encourage you to review this policy proposal and send your comments to <address-policy-wg(a)ripe.net>. Regards, Filiz Yilmaz RIPE NCC Policy Development Officer

10 11

Joking follow-up, second round
by Turchanyi Geza 30 Apr '08

30 Apr '08

Hi, Many thank for your comments in the first round. I appreciated that most of you read my text and commented only after reading as a whole. Those who listened me at RIPE 55 had been in better position. Let me clarify some details: I fully agree that the final document should not use the term "dummy" and "expert". However, these words emphasize the difference between different class of users. 1, Actually an IP address of a DSL user is not in the RIPE database, expect in the very exceptional case when the DSL user have a fix IPv4 address, may be even a subnet. 2, There are many ways to allocate IPv6 network for DSL users; however, it is hopeless to create database entries for them. The reasons are not only technical, but political. With IPv4 most of the DSL users had pseudo-anonym IPv4 addresses, their contact details had been known only by their ISP (and by the police, eventually), and the ISP is not allowed at all to publish the user data in any open database. 3, My suggestion was at RIPE 55 (and in my previous letter): let's do something similar in case of IPv6 what we have done with IPv4. "pseudo" dynamic IPv6! (Which is rather static, BTW.) 4, It is possible to allocate even /48 for every DSL users, even with the "pseudo" dynamic IPv6 allocation mechanism, however, why should we do it? If an "anonym" DSL user could use the same amount of the address space than a RIPE registered DSL user then nobody will register its address space in the database! 5, The IPv6 address space is huge, however, if we do not know who is using it then we will loose it soon. 6, Therefore my suggestion was amends previous policies (what were fine for the "experts", knowing what an IP address is and what a subnet is). 7, I still propose a common policy: how to distinguish between an "anonym" DSL user and a "registered" one. My proposal is: an "anonym" DSL user should receive a minimum IPv6 allocation. In this case the "minimum" is one subnet (/64) or, if there are automatically usable subnets for special purpose then the minimum size might include those subnets. If a DSL user need more then should fill an address request, and registered. 8, I also would prefer if the "anonym" DSL users would share a visible different address space than the registered one. If IANA would reserve a prefix (preferably a /16) for the "anonym" DSL (mobile, CATV, etc) users, then the RIRs could allocate big pools from this prefix to ISPs using different allocation criteria than for the "registered" IPv6 networks. Back to the AS-local IPv4 address space concept: Private address space have to be unique within a routing domain, AS-local address space is unique within the given Autonomous System. (A group of autonomous Systems might share even this address space, but this is the exception, and not the rule.) I do think that the AS-local address pool can be created as a collaborative effort. IANA, ISPs can lease address blocks for this pool. This is not trading, but still a reallocation! Any reallocation policy should allow creation of a common address pool! It is easier to allocate "automatic" and "anonym" IPv6 network for DSL users if we have a big enough, better routable IPv4 address pool, an AS-local address pool. Please read also my proposals: http://www.ripe.net/ripe/meetings/ripe-55/presentations/turchanyi--two-joke… and my presentation: ( http://www.ripe.net/ripe/meetings/ripe-55/presentations/turchanyi-two-jokes… Many thanks for your attention, Geza

1 0

Joking follow-up
by Turchanyi Geza 30 Apr '08

30 Apr '08

Hi, Is there anyone experienced in creating RIPE policy documents AND ready to help me to create a new one? (Or: two new ones). In the previous RIPE meeting I gave a talk ( http://www.ripe.net/ripe/meetings/ripe-55/presentations/turchanyi-two-jokes…) and attached a copy of a draft-RFC ( http://www.ripe.net/ripe/meetings/ripe-55/presentations/turchanyi--two-joke… ) I realized that some of my ideas relate to policies, other relate to technologies. As there is a separate policy working group now, I try to create a policy document as well – and need some help to formalize and may be finalize, correct my ideas. Highlights: 1, AS-local IPv4 address pool creation and maintenance 2, IPv6 address pool and address allocation for dummies Details1 (AS-local IPv4): The drafts ( http://www.ripe.net/ripe/meetings/ripe-55/presentations/turchanyi--two-joke…) gives more details, please find a short summary below. One of our responses to IPv4 address scarcity was the creation of "IPv4 private address pool" in 1994-1996. However: The scope of private addresses is not defined well; The private address pool size is too small for large ISPs; Network Address Translation should be in use at every routing domain borders. AS-local IPv4 pool should be similar but a little bit different compared to private address pool: Uniquely use in every Autonomous system (or collaborative group of ASs) Different set of IPv4 addresses (different scope!); Mechanism to add and revoke address-blocks by contributors to this pool should be implemented (in order to create a contribution-friendly atmosphere); Network Address Translation should be applied only if the destination address is outside of the originator Autonomous System boundary. The introduction of AS-local addresses would help us not only maintain our present IPv4 service, however, support the IPv4->IPv6 transition. (See below) Details2 - IPv6 address pool and address allocation for dummies: As everybody knows, there are well defined IP address allocation policies for fixed, static networks, like an University campus, or an enterprise network. These sites should have administrative and technical contact persons, the "tech" knows what an IP address is, the "admin" pays the bill, and both person is in the database of the Regional Registry. However, a huge part of the IP address space is used differently: both the "tech" and the "admin" work for the ISP, and the actual costumer of the IP address might not even know that he/she is using an IP address. (is a dummy costumer, only in this respect). This is the typical case in DSL environment today with IPv4. The introduction of IPv6 won't change too much. Shall we treat and regulate the IP address allocation for the "dummies" in the same way as we do it for the "experts"? I do not think so. In fact, we can not. Is there any policy for the "dummies"? I was unable to find it. If you have 30 millions "dummy" DSL (or cable modem, or mobile-phone) users how would you provide IP addresses for them? Of course, global addresses are the best. However, as there are not enough global addresses, some tricks should be applied. Common practice: allocate IP addresses dynamically. (BTW: dynamic allocation also mean pseudo-anonym and temporary allocation.) Dynamic allocation saves addresses considerably. However: If only 50% of the costumers connect at peak time today, tomorrow this may increase to 60%. That means: the need for addresses increased 20% while the costumer base is still the same. Using non-global, reusable IP addresses still does not solve all the problems. 30 millions is much more than the total size of the private address pool. Even if the ISP would assume, that not all users connect to the network at the same time, it might not help for long time as the number of costumer being on-line at peak time might increase. AND: using private addresses also means loosing functions. If your computer has a private address, you can not provide any services outside the private address domain (this stops using a couple of popular games, etc) This restriction is unavoidable consequence of using any kind of reusable addresses. However: the private address domain is very restricted. By using AS-local addresses, we would have a larger routing domain and fewer restrictions.) If we create an AS-local address pool, then it is possible to allocate reusable IP addresses in a more stable manner. This allocation is still a dynamic allocation, however, rather stable AND easy to couple IPv6 allocation with it. However, if we allocate IPv6 networks for every costumer that use dynamic IPv4 allocation today then most of them won't use for a while the IPv6 stuff. AND this IPv6 allocation will be pseudo-anonym, not directly reflected in the RIPE (or other RIRs) database. Therefore I suggest that ISP-s should have a dedicated IPv6 address pool for "dynamic IPv6" allocations and these address pool should be easily recognizable. (This was the reason why I proposed in my talk at RIPE 55, that all "dynamic IPv6" pool should be allocated from an IANA dedicated /16 prefix) The size of the "dynamic IPv6" network should be the minimal one: /64. If there are mechanism that allows automatic use a subnet, than a little bit bigger size might be allowed (max /60), however if /56 or /48 would be allowed than there wont be any more interest to have a RIPE registered network instead a "dynamic" one, therefore my suggestion is to declare in the policy that a "dynamic" IPv6 allocation should be as narrow as possible. OK. Please help me to rewrite the above idea to formulate policies. Thanks, Geza Turchanyi INFO-C

6 7

Agenda for RIPE 56 - 1st draft
by Gert Doering 30 Apr '08

30 Apr '08

Hi APWG folks, below you can find a draft for the agenda for the RIPE address policy WG meeting at the upcoming RIPE meeting in Berlin. The APWG meeting will take place Wednesday, 11:00-12:30 and Thursday, 09:00-10:30. If there is anything else you want to see on the agenda, please let me/us know! In the plenary section on Tuesday morning, we will have a presentation from Filiz Yilmaz on ongoing policy work in other regions, so we know what "they" are working on, and can try to learn from it. (Thanks!) The exact time lines depend a bit on the outcome of the policy proposals currently (and soon to be) in "last call". regards, Gert Doering, APWG chair ----------------------- snip ---------------- Wednesday, 11:00-12:30 A. Administrative Matters 5 min (selecting a scribe, approving the minutes, etc.) B. Overview over concluded protocols 10 min B.1 2007-07 - End Policy for IANA IPv4 allocations to RIRs withdrawn, replaced by 2008-03 B.2 2007-06 - Global Policy for the Allocation of the Remaining IPv4 ... withdrawn, replaced by 2008-03 B.3 2005-08 - Proposal to Amend the IPv6 Assignment an Utilisatio ... (change of /48 to "LIR's choice", /56, and HD-Ratio) accepted B.4 2007-03 - IPv4 Countdown Policy withdrawn, replaced by 2007-06/2007-07, replaced by 2008-03 B.5 2007-04 - IANA Policy for Allocation of ASN Blocks to RIRs accepted in RIPE region (short feedback from ICANN/AC on current global status) C. New Proposals since RIPE 55, just presentation, no discussion yet 15 min C.1 2008-01 - Assigning IPv6 PI to Every Inetnum Holder C.2 2008-02 - Assigning IPv6 PA to Every LIR C.3 2008-03 - Global Policy for the Allocation of the Remaining IPv4 Address Space (IPv4 countdown policy, revisited) [C.4 2008-* - if anything else appears in the meantime] D. "Contracts and Certification session" D.1 2007-01 - wrap up, next steps, comments from the NCC board D.2 report from the Certification Authority (CA) TF - update on operational status [Nigel Titley] 15min - life certification demo 20min - discussion: proposal for initial certifications 10min D.3 new policy proposal from the CA TF, based on discussion Thursday, 09:00-11:30 E. Restart PI discussions (if we have consensus on 2007-01) E.1 2006-01 - Provider Independent IPv6 Assignment for End User Organizations E.2 2006-05 - PI Assignment Size E.3 2008-01 - Assigning IPv6 PI to Every Inetnum Holder F. "End of IPv4" session - generic discussion on how to proceed F.1 2007-08 - Enabling Methods for Reallocation of IPv4 Resources ("The Market" proposal) F.3 2007-09 - Cooperative distribution of the end of the IPv4 free pool (Tony Hain's proposal) F.3 2008-03 - Global Policy for the Allocation of the Remaining IPv4 ... (/8 reservations to RIRs) Z. AOB -- Total number of prefixes smaller than registry allocations: 113403 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

1 1

2008-01 Review Period extended until 16 May 2008 (Assigning IPv6 PI to Every Inetnum Holder)
by Filiz Yilmaz 30 Apr '08

30 Apr '08

PDP Number: 2008-01 Assigning IPv6 PI to Every Inetnum Holder Dear Colleagues, The Review Period for the proposal 2008-01 has been extended until 16 May 2008. With the acceptance of this proposal, the RIPE NCC will conduct a one-time operation to assign a /56 IPv6 PI prefix to all End Users with an IPv4 assignment registered in the RIPE Database. You can find the full proposal at: http://www.ripe.net/ripe/policies/proposals/2008-01.html We encourage you to review this policy proposal and send your comments to <address-policy-wg(a)ripe.net>. Regards, Filiz Yilmaz RIPE NCC Policy Development Officer

7 10

2008-03 New Draft Document Published (Global Policy for the Allocation of the Remaining IPv4 Address Space)
by Filiz Yilmaz 30 Apr '08

30 Apr '08

PDP Number: 2008-03 Global Policy for the Allocation of the Remaining IPv4 Address Space Dear Colleagues, We have published a draft document for the proposal described in 2008-03 and the proposal is moved to Review Phase. This proposal describes the process for the allocation of the remaining IPv4 space from IANA to the RIRs. When a minimum amount of available space is reached, one /8 will be allocated from IANA to each RIR, replacing the current IPv4 allocation policy. You can find the full proposal at: http://www.ripe.net/ripe/policies/proposals/2008-03.html and the draft document at: http://www.ripe.net/ripe/draft-documents/2008-03-global-policy-draft.html We encourage you to read the draft document text and send any comments to address-policy-wg(a)ripe.net before 27 May 2008. Regards Filiz Yilmaz RIPE NCC Policy Development Officer

2 1

2007-08 Review Period extended until 28 May 2008 (Enabling Methods for Reallocation of IPv4 Resources)
by Filiz Yilmaz 28 Apr '08

28 Apr '08

PDP Number: 2007-08 Enabling Methods for Reallocation of IPv4 Resources Dear Colleagues, The Review Period for the proposal 2007-08 has been extended until 28 May 2008. This proposal outlines a framework to migrate previously allocated IPv4 resources from one Local Internet Registry (LIR) to another LIR within the RIPE NCC Service Region. You can find the full proposal at: http://www.ripe.net/ripe/policies/proposals/2007-08.html We encourage you to review this policy proposal and send your comments to <address-policy-wg(a)ripe.net>. Regards, Filiz Yilmaz RIPE NCC Policy Development Officer

1 0

RIPE 55: APWG minutes (draft)
by Gert Doering 25 Apr '08

25 Apr '08

Hi APWG readers, there seems to have been a slight coordination problem regarding the posting of the APWG minutes to the mailing list - sorry for that, it seems everyone was assuming "it has already been done", but checking my archives, I can't find any trace of it. Anyway - here are the (draft) minutes from the last APWG meeting at RIPE55 in Amsterdam. Please let us know about everything that's not correct, so that I can declare the minutes as final in Berlin. (And a big thanks to the RIPE NCC folks for creating these truly amazing minutes!) Gert Doering APWG Chair ----- Forwarded message from Susannah Gray <susannah(a)ripe.net> ----- Minutes from RIPE 55 RIPE Meeting: 55 Working Group: Address Policy Status: Draft Revision Number: 0 -------------------------------------------------------------------------------- Wednesday, 24th October 2007 11-12:30 Chair: Gert Doering Minutes: Scott Donald (RIPE NCC) Jabber Scribe: Laura Cobley (RIPE NCC) ----- 1. Administrative Matters * Welcome * Select a scribe - Scott Donald (RIPE NCC) * Finalise agenda * Minutes from RIPE 54 were approved http://www.ripe.net/ripe/wg/address-policy/r54-minutes.html * Co-chair changes Andrea Borgato stepped down as co-chair of this working group. Gert Doering and Sander Steffann remain as the two co-chairs. Speak to one of the co-chairs if you are interested in becoming a co-chair. ----- 2. Overview of Concluded Policy Proposals Proposal 2006-02 "IPv6 Address Allocation and Assignment Policy" - was accepted and implemented Proposal 2006-04 "Contact E-Mail Address Requirements" - was withdrawn Proposal 2006-06 "IPv4 Maximum Allocation Period" - was accepted and implemented Proposal 2006-07 "First Raise in IPv4 Assignment Window Size" - was accepted and implemented Proposal 2007-02 "Change in IP Assignments for Anycasting DNS Policy" - was withdrawn Proposal 2007-03 "IPv4 Countdown Policy" - was withdrawn Proposal 2007-04 "IANA Policy for Allocation of ASN Blocks to RIRs" - was accepted by RIPE, and is awaiting other RIRs acceptance ----- 3. Generic Discussion on PI Session Presentation: Feedback from the RIPE NCC Executive Board Regarding Proposal 2007-01 - Axel Pawlik (RIPE NCC) http://www.ripe.net/ripe/meetings/ripe-55/presentations/pawlik-2007-01.pdf There were no questions Presentation: Proposal 2007-01 "What do we do next?" - Nick Hilliard (INEX) http://www.ripe.net/ripe/meetings/ripe-55/presentations/hilliard-2007-01.pdf Gert Doering: commented yearly price of a PI assignment could be 100-300 Euros per year. More than that and the RIPE NCC would make a surplus. The Board would decide the cost. Wilfrid Woeber (ACONET): said that Axel gave some alternative proposals but asked if this proposal was talked about/agreed by the Board? Axel Pawlik (RIPE NCC): said this proposal would make things easier for us. It puts some responsibility on members to help maintain the link with the PI holders. He thought the board would be more positive about this than the previous suggestions. The general meeting, perhaps in May, can discuss changes to the charging scheme. Wilfrid Woeber: asked from the data privacy point of view, what would happen if an end user asks for his information to be removed. From this proposal the resources would eventually be reclaimed. We would need to think about solutions to that. Nick Hilliard: said the RIPE NCC may be breaking data privacy laws for NOT maintaining accurate information on the RIPE database at the moment. Daniel Karrenberg (RIPE NCC): thanked Nick for trying to fix a problem that other people saw, and to put things right again. We tried to encourage PA and not PI in the past (since RIPE 127) but somehow PI became free of charge. So we need to correct this situation. We need to do a lot of clearing up of the old data. Todd Underwood (renesys/babbledog): said that with the DNS system if you stop paying your domain name stops resolving. But with Whois information, nothing would happen if you stopped paying and your details were removed. Nick Hilliard: said RIPE are not the police, they are clearing house and numbers registry. RIPE guarantees uniqueness of the internet resources it issues. If someone hijacks your IP space you need to take it up with the ISP and solve it yourself. Todd Underwood: said the only thing RIPE can do with an expired registration is give it to someone else, or leave it unused in this proposal. So the analogy with the DNS system is flawed. Ruedi Volk (Deutsche Telekom): said clearing up the Whois would also clean out old route objects which are referencing the IP space. Nick Hilliard: said if you stop paying you will lose the entitlement to use it. So you would lose the unique access to this space. Leo Vegoda (IANA): Supported the proposal and wanted implementation speeded up as there are thousands of entries to contact and we want to encourage light regulation of the industry (not heavy regulation). Nick Hilliard: said if consensus is reached we can start quickly. The next opportunity to implement it is May 2008 due to the budgetary impact. Leo Vegoda: agreed it must be approved by the RIPE community. Asked to keep it as simple as possible, and refinements can be made later if required. Nick Hilliard: proposed to put 2007-01 back on the address policy working group for further discussions. Daniel Karrenberg: said he saw no one against the proposal. People not updating their contact details would not be cut off if they are using the IP space. The RIPE NCC has tools (like RIS) that can see if IP space is being used. Perhaps we should add something to the policy saying that LIRs should hold information on the organisations they are routing in case we need to contact them. Nick Hilliard: said that due to data protection, all we can do is ask the LIR to contact the end user for permission to give their contact details to the RIPE NCC. We can't compel LIRs to supply this information. Daniel Karrenberg: said that end users can be encouraged to supply correct contact information to LIRs so that the LIR can update the RIPE Database. The RIPE NCC needs something in writing to implement this. Wilfrid Woeber: wanted the community to give the RIPE NCC a mandate in order to give some sticks to motivate people to collaborate. Another stick is certification, or giving the NCC the right to update objects on the RIPE Database. Another point is that PI is not free, it counts towards the billing score of an LIR for the first 12 months only. This was based on RIPE NCC not charging for holding resources. But now we see there is a running cost for holding and tracking resources so the charging scheme needs to be looked at. PI is too cheap but not free. Hans Petter Holen (Visma IT): thought it was a good proposal. We need a stronger link between the database and who is holding the resources, which is why charging and some kind of certification is needed in the long term. We need to get approval from the community on the proposal and budgetary stuff, then decide how we will police this. LIRs can filter on information on the RIPE Database for example, also black holes could be be used for people who don't pay their bills. Carsten Schiefner (Deutsche Telekom): liked the proposal. Would like more of the policing function in the proposal. Also would like to see what it would cost to implement what Hans Petter Holen suggested. Richard Cox (Spamhaus project): has investigated inappropriate bgp announcements- whatever addresses are used by people doing bad things are bogus. The only solution to get any sort of policing is to get the Tier1s on board. LIRs would need to have contractual agreements with end users saying that only they can use the IP space assigned. Have we asked them? Nick Hilliard: said this is a difficult problem- it involves creating access lists for starters. A lot of Tier1s and Tier2s filter customer announcements already. Inappropriate bgp announcements falls outside the scope of this proposal. Denis Walker (RIPE NCC): supported this proposal. The RIPE NCC is legally obliged to make sure the data on the RIPE Database is accurate so we want a mandate to clean up the database. We receive complaints sometimes due to problems originating from IP space that has incorrect contact details. But we are unable to do anything about it. Gert Doering: said he heard only supporting voices and asked for show of hands if anyone opposed this proposal- no hands went up. Action point on RIPE NCC: prepare for the implementation of Nick Hilliard's proposal if approved by the community in the Address Policy Working Group. Action point on Nick Hilliard: put proposal forward on the mailing list. ----- 4. Presentation : Proposal for Restructuring the IPv6 Address Allocation and Assignment Policy - Leo Vegoda (personal view) http://www.ripe.net/ripe/meetings/ripe-55/presentations/vegoda-v6-policy.pdf Gert Doering: said that we got rid of the 200 rule for IPv6 allocations. George Michaelson (APNIC): said people don't usually come back for more IPv6 space. We are giving out very large amounts of IPv6 space at the moment. Geoff Huston (APNIC): said the PI/PA distinction is confusing everyone and removing it would be good, because an address is an address. We need to decide what we want to achieve from the proposal and what the industry needs from us. Do we want LIRs to get an IPv6 allocation and never come back, or come back for an allocation after a certain time period (like one year). We don't want to replicate IPv4 policies which ended up in scarcity. However coming back for IPv6 allocations every year is maybe too much. Gert Doering: said a PI is given for a well defined need for a well defined network. A PA block allows for growth which is difficult to plan. Coming back every year/ 6 months for new allocations for example gives aggregation problems. Ruedi Volk: said scarcity in IPv6 is no problem, but the routing table size is. The only way to keep control of the number of routes is the PI/PA distinction. Cathy Aronson (ARIN): said one big distinction between PA and PI is that you can't easily change providers with PA. -------------------------------------------------------------------------------- Thursday, October 25, 09:00-10:30 5. Discussion of All Other Current Open Policy Proposals Proposal 2005-08 to Amend the IPv6 Assignment & Utilisation Requirement Policy Sander Steffann: consensus has been reached on this proposal and it will be going through. Proposal 2007-05 IPv6 ULA-Central Sander Steffann: this is waiting for the decision by the IETF. We will re-start this proposal if the IETF gives the go-ahead. If not approved by the IETF it will be withdrawn. ------ 6. End of IPv4 Session Presentation: Proposal 2007-06 "Global Policy for the Allocation of the Remaining IPv4 Address Space" - Roque Gagliano (ANTEL) http://www.ripe.net/ripe/policies/proposals/2007-06.html Toshiyuki Hosaka (JPNIC): also has proposal 2007-07 which is the same as the proposal above except it says only a single /8 will be given to each RIR at the point when new IANA free pool hits 5 */8 Wilfrid Woeber: said it is good to make things reasonable equal at the end. To avoid RIR shopping the number of allocations (N) should be small (1 or 2). Nick Hilliard: said fairest way is N=0. Burn rates in each RIR are significantly different, for example AFRINIC and APNIC, which could create political problems and horse trading later. Addresses should be allocated as needed not for any political purpose. Roque Gagliano: said it is possible that AFRINIC could get the last /8 for example depending on which RIR qualified for an allocation at a particular time. Jordi Palet (Consulintel): said he was against a policy change. Whatever we try to do to be fair may not work, as the policy could be changed again in a few months anyway. One of the aims of the proposal may be to give developing regions more time to implement IPv6, but this could damage the implementation of IPv6 in these regions. Ruedi Volk: said that only N=0 or 1 is acceptable. 1 gives a predictable end game. But the period before this will still be unpredictable which cannot be avoided. Raul Echeberria (LACNIC): said he has a lot of sympathy with the proposal. The current situation is against the conservation of IPs because the more you allocate the more you get. Certainty and predictability is important at the end. Gert Doering: said this proposal will move the uncertainty phase sooner. Another suggestion is to make IPs more difficult to get- we should never do that because IPs should be issued on the basis of need. Roque Gagliano: said other RIRs gave feedback saying this proposal will allow more time for planning. If the gentleman's agreement about 2 * /8s at a time is not stuck to, one of the big RIRs could get the last 8 * /8s for example. Geza Turchanyi (Magyar Telekom): said we should go to IPv6 faster to avoid this solution. The RIR receiving the last /8 will have a lot more IPs than the other RIRs have. Niall Murphy (Google): said RIR shopping should be considered in the solution. Roque Gagliano: said we don't know what will happen after exhaustion day. Some large multinational companies can already get addresses from different RIRs around the World. Niall Murphy: said some large organisations with a presence around the World will be able to RIR shopping, but smaller ones will not. N=1 or 0 looks okay, but 5 is odd. David Kessens (Nokia Siemens): said we are in the business of giving addresses to people that need it, not to reduce uncertainty for RIRs. Carsten Schiefner: said certainty argument not so important. The IP needs for larger RIRs are greater than that for smaller RIRs. Roque Gagliano: said we continue with current policies but we are only talking about the last /8s. Any formula we decide on, someone can consider unfair. Laura Cobley (RIPE NCC): Jabber comment from Per Heldahl- no matter what we do there will always be someone who gets the last /8. He thinks all the RIRs should run out at approximately the same time. He suggest that IANA could issue single /8s at a time. Gert Doering: said that was an interesting suggestion. He thought that saying the RIRs would not stick to the gentleman's agreement is somewhat offensive as it shows some mistrust. Geoff Huston: said the end of the free allocations does not mean the end of IPv4. We should be worrying about what happens next instead of this because there is not much time left. Gert Doering: summary of discussion- no support for N=5, some support for N=1. A lot of voices for 0. No consensus in the room. If we do it, N=1 is the one to pursue. We will discuss in the working group and the PDP process. Wilfrid Woeber: asked if N=5 was ratified by APNIC? Roque Gagliano: said it is waiting for the LACNIC board's decision. Wilfrid Woeber: said this will need global agreement. N=5 won't be agreed on globally. Raul Echeberria: said he thinks the board will not approve this so the discussion remains open. ----- Presentation: Proposal 2007-08 Reallocation of v4 resources - Remco van Mook (Virtu) http://www.ripe.net/ripe/meetings/ripe-55/presentations/van-mook-2007-08_re… Geoff Huston: asked if there will be any conditions or constraints on the LIR "disposing" of an allocation. In the APNIC proposal the "disposing" LIR can't get another allocation for 2 years. 2 years because it will take 1 year for the proposal to go through and APNIC will probably not be able to issue more allocations 2 years after that. So the selling LIR can't come back and get another allocation to sell. Remco van Mook: thought restraints like this would decrease liquidity of IP addresses. We would like people to start using this proposal before we run out of addresses. Geoff Huston: said in the APNIC proposal the acquirer of the allocation would be required to have a 80% usage rate on all its allocations (including acquired ones) before it could qualify for an additional allocation. Remco van Mook: said it is not in this proposal, but usage rates will be looked at under current policies. Gert Doering: said that 80% usage to qualify for an additional allocation should be made clear. Geoff Huston: said people might acquire IPs to sell them later. An idea he heard might be that allocations can only be transferred once, to try and reduce the number of speculators. Remco van Mook: said it would be bad for liquidity to only be able to transfer an allocation once. Leo Vegoda: said the proposal says allocations can only be transferred within the RIPE region. Is it an issue if the physical address of buying organisation is outside the region? Remco van Mook: said it would not be a problem as long as the address space is registered with the RIPE NCC. Laura Cobley (RIPE NCC): Jabber comment from Per Heldahl- NRO guidelines say that IPs are not held as property. These guidelines would need to be changed. Gert Doering: said the guidelines say addresses are not property but companies are being bought with networks and addresses included. Geza Turchanyi: said we should create a pool for re-using addresses. Hans Petter Holen: said people are already buying and selling companies which include IPs. Today we assign this commodity based on need. We are creating a market for this commodity. Maybe we need to do some research into what happened in other industries in the past in these situations. Cathy Aronson (ARIN): said perhaps we should have a restriction such as you are not allowed to sell IPs until 2 years after they were allocated. Remco van Mook: said that was a good idea. Mark McFadden (BT): said RIPE already has a policy and mechanism for transferring allocations between LIRs. He is worried about the unintended consequences of this proposal which are that IPs can be seen as assets so people may be taxed on them. The market part is worrying. Remco van Mook: said we should seek some advice from economic experts outside the community. Gert Doering: summary of discussion- some support and some words of caution. We need to discuss further on the mailing list. ----- 7. Formulation of an official statement from the RIPE community regarding the end of IPv4, and recommendations for moving towards IPv6. - Sander Steffann http://www.ripe.net/ripe/meetings/ripe-55/presentations/steffann-ripe-resol… Gert Doering: said external organisations have asked what our official opinion is on IPv6. As the RIPE NCC is a secretariat they can't make political statements which is why we are asking you. It will be finalised in the IPv6 working group. Jim Reid (RIPE NCC executive board): said point 5 should mention software vendors in addition to equipment vendors. Rob Blokzijl (RIPE): said this statement is meant for an external audience, like governments and policy makers. It is not a manual about how to implement IPv6. We should focus on the message and not have too much detail, or the intended audience will lose interest. Sander Steffann: said the idea is to give people a message to move to IPv6. Geza Turchanyi: said the statement that the internet will continue as normal is too strong. Jim Reid: said the message should be business as usual even if we run out of IPv4 addresses. Websites and email etc will still work after IPv4 runs out. Sander Steffann: asked for feedback on point 2. Ruedi Volk: asked if this statement is saying application providers don't have to do IPv6 until network providers and ISPs do the gateway? Sander Steffann: said maybe the wording needs to be broadened. Ruedi Volk: said we should cut down specifics as much as possible. Sander Steffann: asked for feedback on point 3. Bernard Tuy: said it is not useful to have this point here. IPv6 has been in operation for years. Rob Blokzijl: said this point will confuse the audience. It should convey the message that we don't need other organisations involvement to deal with this problem. It needs re-wording. Sander Steffann: agreed some re-wording is needed. He then asked for feedback on point 4- there were no comments. He then asked for feedback on point 5. Rob Blokzijl: said drop this as it is covered by point 4. Niall Murphy: said pressure on vendors comes in tender documents, not here. Remco van Mook: said keep the message as short as possible, the message is do IPv6 now. Geza Turchanyi: said IPv6 is running in the backbone but for DSL is poor. Some encouragement for IPv6 in this area should be encouraged. Jord Palet: said a message to end users to request IPv6 is not here. You could make separate texts for users, vendors, policy makers. The message should be focussed not short. Rob Blokzijl: said he does not agree with Jordi. The community needs a single short statement for policy makers, governments etc. There is a problem (with IPv4 exhaustion) but not as bad as people think. Sub-sections will not be useful. David Kessens: said the message needs to be short and simple. This will be discussed in the IPv6 Working Group, where a longer session has been made available. Gert Doering: summary of discussion- Sander Steffann and I will work with Rob Blokzijl and David Kessens in the working group to try and finalise this. END---- ----- End forwarded message ----- Gert Doering -- NetMaster -- Total number of prefixes smaller than registry allocations: 110584 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

1 0

Does any kind of IPv4 transfer policy make sense
by michael.dillon＠bt.com 24 Apr '08

24 Apr '08

I've been thinking a lot about the various proposals for new IPv4 transfer policies, many of which seem to want to create a market for IPv4 address blocks. In a recent discussion, someone suggested that there are two things that RIRs can and should do, in order to mitigate the effects of IPv4 exhaustion. 1. Reclaim unused allocations 2. Allow IPv4 address block transfers In most of the discussions about IPv4 exhaustion, these two actions are treated as separate activities. The first is an extension of what the RIRs and IANA have already been doing for many years. And the second is a wonderful new idea that will let organizations buy IPv4 addresses when RIPE runs out. But if you look closely and what "transfer" means, it becomes complex, confusing, and very unclear. In my opinion, is transfer is not separate from reclamation. In fact it *IS* reclamation. Instead of RIPE looking for unused allocations, it is LIRs looking for unused assignments. What happens after the unused addresses are identified is really not as important as the fact that no reclamation/transfer can happen until unused addresses are found. How likely is it that we will find unused address blocks at a time when the fresh supply of IPv4 addresses has run out? The RIRs have always tried to have transparent processes surrounding IP addressing. That's why we have the RIPE database which publishes a directory of who uses any particular IP address range. But how accurate is it. How many of these IP addresses are really in use? As long as we assume that the RIPE database is basically accurate, and that a higher percentage of all IPv4 addresses will be in a "used" state when we reach IPv4 exhaustion, this implies that there will be very few IP address blocks available to be transferred. And if that is true, this implies that the price per address block (if selling is allowed) will be very high, and rise over time as the unused addresses in the RIPE database become more scarce. Those prices could very well rise to such a high level that it is cheaper to simply buy out smaller companies and disconnect less profitable customers in order to get access to their address blocks. I feel that we need to pay more attention to analysing what addresses are really in use (used in source/destination headers of IP packets on the Internet) and to providing support and education to LIRs so that they can better audit their own allocations. --Michael Dillon

2 1