June 2015 - address-policy-wg

2014-03 New Draft Document and Impact Analysis Published (Remove Multihoming Requirement for AS Number Assignments)
by Marco Schmidt 07 Nov '15

07 Nov '15

Dear colleagues, The draft document for version 2.0 of the policy proposal 2014-03, "Remove Multihoming Requirement for AS Number Assignments" has now been published, along with an impact analysis conducted by the RIPE NCC. Some of the differences from version 1.0 include: - AS Numbers can now be assigned for any need that cannot be satisfied with an existing AS Number - The RIPE NCC's need evaluation has been removed - A limit of 1,000 AS Numbers per organisation has been added - There is now a requirement that 16-bit AS Numbers be multihomed after nine months - The requirement that the routing policy must be new and unique has been removed - Related wording adjustments in the summary and rationale of the proposal You can find the full proposal and the impact analysis at: https://www.ripe.net/ripe/policies/proposals/2014-03 And the draft document at: https://www.ripe.net/ripe/policies/proposals/2014-03/draft We encourage you to read the draft document text and send any comments to <address-policy-wg(a)ripe.net> before 22 January 2015. Regards Marco Schmidt Policy Development Officer RIPE NCC

16 41

2015-02 Draft Document and Impact Analysis Published (Keep IPv6 PI When Requesting IPv6 Allocation)
by Marco Schmidt 07 Jul '15

07 Jul '15

Dear colleagues, The draft document for the proposal described in 2015-02, "Keep IPv6 PI When Requesting IPv6 Allocation" has been published. The impact analysis that was conducted for this proposal has also been published. You can find the full proposal and the impact analysis at: https://www.ripe.net/participate/policies/proposals/2015-02 and the draft document at: https://www.ripe.net/participate/policies/proposals/2015-02/draft We encourage you to read the draft document text and send any comments to address-policy-wg(a)ripe.net before 7 July 2015. Regards, Marco Schmidt Policy Development Officer RIPE NCC

22 21

"ALLOCATED UNSPECIFIED"
by Enno Rey 02 Jul '15

02 Jul '15

Hi, some of you might already cringe just from this mail's subject ;-) I'm currently involved in handling some netblocks which are in "ALLOCATED UNSPECIFIED" state and this turns out to be surprisingly difficult, even in cases where both organizations (that is the LIR holding the covering aggregate and the organization which received the "more specific" PI assignment back in the 90s) apparently agree on a course of action. My impression is that these difficulties not least arise as seemingly no policy exists on "how to convert those assignments into 'ASSIGNED PI' or 'ALLOCATED PA' space". I'm aware that these netblocks might only be "remnant corner cases" totally irrelevant to the majority of the community. Which brings me to the following questions: a) do any of you "feel the same pain" when it comes to these blocks? b) do you think a policy proposal should be prepared how to handle those? I'm willing to prepare sth. c) what could such a proposal look like? What do those concerned think how a reasonable way of moving those blocks into a "stable state" can be identified/described. many thanks in advance for any type of feedback. everybody have a pleasant evening Enno -- Enno Rey ERNW GmbH - Carl-Bosch-Str. 4 - 69115 Heidelberg - www.ernw.de Tel. +49 6221 480390 - Fax 6221 419008 - Cell +49 173 6745902 Handelsregister Mannheim: HRB 337135 Geschaeftsfuehrer: Enno Rey ======================================================= Blog: www.insinuator.net || Conference: www.troopers.de Twitter: @Enno_Insinuator =======================================================

2 1

We need IPv4 transfers
by Shahin Gharghi 30 Jun '15

30 Jun '15

Hi Although this is the last call, and I have been criticizing the usefulness of this proposal in protecting the remained IPv4, I am not still logically replied. Those questions were: 1- How can we prevent transfers by accepting this proposal? We can always transfer by taking ownership of the sellers company. So this proposal would not be beneficial at all. 2- According to the 1st question, and knowing that , there are some dealers who get IP's from RIPE NCC and sell them to customers, Why wouldn’t RIPE NCC sell the IP's directly to them? (By allowing those companies to register new LIR and get new /22). 3- Obviously the internet is increasing and companies need more IP's, and there are some other IP's available in RIPE NCC, Why shouldn't we use them? The people won't think about IPv6 seriously unless they see there is no other IPv4. 4- If people are unable to transfer IP's. They will lend them. (The proposal won't help again) I would be glad to add my arguments to Sander's conclusion. -- Shahin Gharghi

1 0

Re: [address-policy-wg] We need IPv4 transfers
by Manager 30 Jun '15

30 Jun '15

Agree, propousal is not good. There ara lot of facts about it. If propousal will be accepted it s good idea to publish current stats at public and compare situation after 6, 12, 24 months. Becouse free ipv4 at ripe is growing. It's better to make ipv4 easy to get. Lets say give new lir /21 but not /22. Then a lot of companies who has ip space not in use prefer to transfer it. And space will be more fair redistributed between members. And any limits on transfers are not good too. Then situation will be controled by members, but not ripe. Yuri@ip4market Отправлено с устройства Samsung -------- Исходное сообщение -------- От: Shahin Gharghi <shahin(a)gharghi.ir> Дата: 29.06.2015 22:16 (GMT+02:00) Кому: address-policy-wg(a)ripe.net Тема: [address-policy-wg] We need IPv4 transfers Dear group members Hi ! I am seriously against this proposal, because it won't be able to stop making financial profit out of IPv4. First of all, as mentioned in proposal, by this rate of transfers, last /8 will be exhausted in 5.5 years and we will have to switch to IPv6 in 5 years… So do you think that we should still care about it?! Knowing that these IP's are valuable, I think that they must belong to RIPE NCC's members and it is their right to have them, so We should make a policy to increase the startup fees or get new entrants , to pay for the whole year fee or transfer fee, and this way it helps us reduce the annual fee of other LIR's and prevent from the unnecessary transfers. On the other hand, it should be considered, that if people are not be able to sell the IP's, they will lend them to others with high rates and therefore they can make financial profits again. Above all, you know that, we have reserved a lot of IP's before ( Multicast, Class E, private ranges and etc.) and I think, it should not be important to reserve more and more again…. They are supposed to be used oneday, since no one would need IPv4 after implementing IPv6. The people, who are in need of IP’s would have no other choice to buy, I would like to know the proposal’s solution for this problem? The most abusers have gotten IP's before 2012 and have already sold most of them, moreover, there is a bug, you can transfer the IP's by taking ownership of the whole organization and I found these transfers useful, because anyone who needs IP would be able to buy it, otherwise how RIPE NCC can fairly distribute IP's ? In the end, I would like to ask those LIR's that have recently joined us and need more IP, not to be shy and courageously tell us about their opinions. -- Shahin Gharghi

3 2

We need IPv4 transfers
by Shahin Gharghi 30 Jun '15

30 Jun '15

Dear group members Hi ! I am seriously against this proposal, because it won't be able to stop making financial profit out of IPv4. First of all, as mentioned in proposal, by this rate of transfers, last /8 will be exhausted in 5.5 years and we will have to switch to IPv6 in 5 years… So do you think that we should still care about it?! Knowing that these IP's are valuable, I think that they must belong to RIPE NCC's members and it is their right to have them, so We should make a policy to increase the startup fees or get new entrants , to pay for the whole year fee or transfer fee, and this way it helps us reduce the annual fee of other LIR's and prevent from the unnecessary transfers. On the other hand, it should be considered, that if people are not be able to sell the IP's, they will lend them to others with high rates and therefore they can make financial profits again. Above all, you know that, we have reserved a lot of IP's before ( Multicast, Class E, private ranges and etc.) and I think, it should not be important to reserve more and more again…. They are supposed to be used oneday, since no one would need IPv4 after implementing IPv6. The people, who are in need of IP’s would have no other choice to buy, I would like to know the proposal’s solution for this problem? The most abusers have gotten IP's before 2012 and have already sold most of them, moreover, there is a bug, you can transfer the IP's by taking ownership of the whole organization and I found these transfers useful, because anyone who needs IP would be able to buy it, otherwise how RIPE NCC can fairly distribute IP's ? In the end, I would like to ask those LIR's that have recently joined us and need more IP, not to be shy and courageously tell us about their opinions. -- Shahin Gharghi

2 1

We need IPv4 transfers
by Shahin Gharghi 30 Jun '15

30 Jun '15

Dear group members Hi ! I am seriously against this proposal, because it won't be able to stop making financial profit out of IPv4. First of all, as mentioned in proposal, by this rate of transfers, last /8 will be exhausted in 5.5 years and we will have to switch to IPv6 in 5 years… So do you think that we should still care about it?! Knowing that these IP's are valuable, I think that they must belong to RIPE NCC's members and it is their right to have them, so We should make a policy to increase the startup fees or get new entrants , to pay for the whole year fee or transfer fee, and this way it helps us reduce the annual fee of other LIR's and prevent from the unnecessary transfers. On the other hand, it should be considered, that if people are not be able to sell the IP's, they will lend them to others with high rates and therefore they can make financial profits again. Above all, you know that, we have reserved a lot of IP's before ( Multicast, Class E, private ranges and etc.) and I think, it should not be important to reserve more and more again…. They are supposed to be used oneday, since no one would need IPv4 after implementing IPv6. The people, who are in need of IP’s would have no other choice to buy, I would like to know the proposal’s solution for this problem? The most abusers have gotten IP's before 2012 and have already sold most of them, moreover, there is a bug, you can transfer the IP's by taking ownership of the whole organization and I found these transfers useful, because anyone who needs IP would be able to buy it, otherwise how RIPE NCC can fairly distribute IP's ? In the end, I would like to ask those LIR's that have recently joined us and need more IP, not to be shy and courageously tell us about their opinions. -- Shahin Gharghi

12 27

IPv6 PI assignment policy
by Thomas Drewermann 27 Jun '15

27 Jun '15

Dear colleagues, we recently requested an IPv6 assingment on behalf of one Freifunk Community in germany. They wanted to be indipendent from us and to take routing decisions on their own. As they are a freifunk community some of the PI assigment would be used to lease addresses to clients/users. According to NCC the policy currently doesn't permit usage of PI space. Small Hotspot providers and especially Freifunk communities typically can not afford a LIR Membership to be independent. In my opinion the current policy makes it hard to adopt IPv6 in such cases. I'd like to propose a change of the policy to allow PI addresses to be used for clients which don't belong the assigment-holder. This clients are connecting to networks which use address space of the holders PI assignment e.g. via wifi. The difference between an assignment is that there is a single address provided to walk in wifi users rather than a whole subnet delegated for usage by the connecting client/user/customer. How do you think about that situation? What would be your thoughts on such a proposal? Regards Thomas Drewermann Freifunk Rheinland e.V.

9 17

What's going on
by Petr Umelov 24 Jun '15

24 Jun '15

2015-06-24 14:37 GMT+03:00 Marco Schmidt <mschmidt(a)ripe.net>: > Dear colleagues, > > Please find below the monthly overview of open policy proposals and the > stage each has reached in the RIPE Policy Development Process (PDP). > > If you wish to join the discussion about a particular proposal, please do so > on the relevant working group mailing list. > PROPOSAL: 2015-01, "Alignment of Transfer Requirements for IPv4 Allocations" > OVERVIEW: Currently, IPv4 allocations received from another LIR can only be > re-allocated after 24 months, while IPv4 allocations made by the RIPE NCC > can be transferred immediately. This proposal aims to align the transfer > requirements with a 24-month holding period for all IPv4 allocations. > RIPE NCC IMPACT ANALYSIS: Includes the point that currently around 10% of > the allocations made by the RIPE NCC are transferred shortly after they have > been allocated. But there were several calculations showing that the part of the transfers from the last /8 is less than 3% There are facts but europeens somehow don't see them. > STATUS: Last Call > WHERE TO COMMENT: Address Policy Working Group: address-policy-wg(a)ripe.net > DEADLINE: 21 July 2015 > FULL PROPOSAL: https://www.ripe.net/participate/policies/proposals/2015-01 > > > We look forward to your involvement in the PDP. > > Kind regards, > > Marco Schmidt > RIPE Policy Development Officer > RIPE NCC -- Kind regards, Petr Umelov

1 1

Consensus on 2015-01
by Sander Steffann 22 Jun '15

22 Jun '15

Hello working group, Here is your chair's (singular, Gert has abstained from judging consensus as he became too involved in the discussion on the mailing list and might be seen as non-neutral on this policy proposal) analysis on the review phase of RIPE policy proposal 2015-01. At the end of the review phase there was a sudden flood of messages both supporting and opposing the policy proposal. Many of these messages were on or after the deadline: the end of the review phase. As those messages didn't bring forward any new arguments they didn't influence my decision making process. I have included them in this overview for completeness' sake. First the people supporting this policy proposal. There were many people who supported the proposal based on the rationale given in the proposal itself (also known as "+1" messages). Others also stated the reasons why they supported the proposal. These included: - It aligns with original intent (make assignments) of the final /8 policy - It makes it less profitable to overtly act against the original intent of the final /8 policy - It is a good step in the right direction, we may need more steps later Here is a list of people that supported this policy proposal: - Andre Keller - Andreas Larsen (after deadline) - Carsten Brückner (after deadline) - Carsten Schiefner - Christopher Kunz - Daniel Suchy - Dimitri I Sidelnikov - Erik Bais - Florian Bauhaus - Garry Glendown - Gerald Krause - Havard Eidnes - Herve Clement - Jan Ingvoldstad - Jens Ott - Marius Catrangiu - Martin Millnert (after deadline) - Mick O Donovan - Ondřej Caletka - Radu-Adrian FEURDEAN - Riccardo Gori - Richard Hartmann - Robert Sleigh - Sebastian Wiesinger - Thomas Schallar - Tim Kleefass - Tom Smyth (after deadline) - Tore Anderson - Torunn Narvestad (after deadline) - Vladislav Potapov David Freedman asked for clarifications about the impact on the Mergers and Acquisitions procedure of the RIPE NCC. These were answered by Marco Schmidt. Daniel Baeza and Richard Hartmann asked for clarifications on how this policy would be applied to allocations made in the past. Marco Schmidt explained that if accepted this policy would only impact transfers happening after the policy was implemented. Transfers that happened in the past would not be impacted. The policy would be applied to existing allocations though. Allocations made in the past would not be transferrable until they were at least 24 month old. For some people this was a problem as they considered it unfair to those LIRs that had started in the last 24 months with the expectation that they would be able to transfer their allocation from the final /8. The people opposing this policy proposal because they consider it a retroactive change are: - Sascha Luck - Storch Matei - Vladimir Andreev There were many messages on this topic. We consider this objection handled because this policy doesn't actually change anything that happened in the past. This policy proposal is about the requirements of transfers. If this proposal gets accepted transfers that have already happened stay happened, and transfers that are about to happen will be checked against the current policy at that time. This is how RIPE policies have always been applied and this policy proposal is no different. There was a message stating opposition to the proposal by Arash Naderpour, but as no reasons against the proposal were given there is not much we can do with this. Consensus based policy development means trying to address objections until the reasons for the objections are taken away. When no reasons are given this is not possible. Therefore this opposition will not have much weight in my analysis. There was also opposition because people felt that this policy proposal didn't solve a real problem and/or wasn't solving all problems related to abuse of the current final /8 policy. They were: - Amir Mohsen (after deadline) - Aleksey Bulgakov - Arash Naderpour (after deadline) - Borhan Habibi - Ciprian Nica - Olga @ip4market.ru (after deadline) - Petr Umelov - Sergey Stecenko - Storch Matei - Yuri @ntx.ru (after deadline) During the discussion it was shown that the number of transfers from the final /8 pool was increasing, especially for very "young" prefixes. This shows that there this policy does solve a real problem. As with all policy proposals it is clear that one policy proposal will not solve all the potential problems all at once. That there are still other potential problems related to the final /8 policy is noted as an encouragement for future policy proposal authors. There were also people objecting because preventing organisations to open a new LIR and then transfer its address space would mean that the membership growth of the RIPE NCC would be a bit lower, and because the RIPE NCC is funded by its members the lower membership numbers might cause the membership cost per member to increase. These were: - Ciprian Nica - Sergey Stecenko - Storch Matei - Vladimir Andreev The impact analysis by the RIPE NCC however explicitly mentions that "Considering the overall size of the membership, the RIPE NCC does not anticipate a significant impact will be caused if this proposal is implemented.". Finally, there were also objections that the final /8 pool was too big and/or not running out fast enough. This objection was made by: - Ciprian Nica - Storch Matei In the impact analysis however mentions that the current pool will last 5.5 years based on the allocation rate of the last 6 months (up to the writing of the impact analysis). That lifetime may be reduced significantly however if new LIRs continue to join in ever-larger numbers and /22 transfers from last /8 also gain more popularity. As the remaining lifetime of the IPv4 internet is extremely likely to be longer than 5.5 years the lifetime of the final /8 pool seems short as it is. Based on the feedback I see strong support for this policy proposal. All objections seem to be addressed as well, so I hereby declare rough consensus on policy proposal 2015-01 and ask our friendly RIPE NCC Policy Development Officer to move this policy proposal to the Last Call phase. Sincerely, Sander Steffann APWG co-chair

1 2