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

Re: [address-policy-wg] Summary of the PI Task Force's recent discussions
by Wilfried Woeber, UniVie/ACOnet 26 Aug '03

26 Aug '03

>Hay, > >On Thu, Aug 21, 2003 at 02:42:05AM +0200, Daniel Roesen wrote: >> On Mon, Aug 11, 2003 at 08:19:46AM +0200, leo vegoda wrote: >> > Below is a summary of the recent discussions on the PI TF mailing list. >> >> As the pi-tf mailing list seems to have been closed (at least I'm unable >> to subscribe - pi-tf-request(a)ripe.net bounces as unknown), I guess >> further discussion should take place here (which makes some sense). > >Closed, yes. I don't think so, pls. see below. Wilfried. ______________________________________________________________________ From: Mally Mclane <mally(a)ripe.net> To: pi-tf(a)ripe.net Subject: [pi-tf] This list has moved to mailman Date: Tue, 26 Aug 2003 09:26:26 +0200 Dear Colleagues, We have moved this list to mailman, in an attempt to cut down on the spam that was entering it. The list will function exactly as before, and the archives will still be available in the same space. You can edit your personal options at: http://www.ripe.net/mailman/listinfo/pi-tf If you notice any oddities, please contact <ops(a)ripe.net>. Regards, Mally Mclane RIPE NCC - Operations --------------------------------------------------------------------------------

4 3

Summary of the PI Task Force's recent discussions
by leo vegoda 26 Aug '03

26 Aug '03

Dear Colleagues, Below is a summary of the recent discussions on the PI TF mailing list. The policy for portable address space is on the draft agenda for the Address Policy WG's session at RIPE 46 on 3 September. Kind regards, -- leo vegoda RIPE NCC Registration Services Manager A summary of the PI TF's initial discussions was agreed and posted to the <lir-wg(a)ripe.net> mailing list on 6 May 2003. It can be found at: http://www.ripe.net/ripe/mail-archives/lir-wg/2003/msg00265.html Gert Döring presented on the discussion at the RIPE 45 LIR WG session. His slides can be found at: http://www.ripe.net/ripe/meetings/ripe-45/presentations/ripe45-lir-pi-tf. pdf At RIPE 45 there was an agreement that the Task Force should continue its work. There was also an agreement that the PI policy is tied to the qualifying criteria for an initial IPv4 allocation. Discussion focused around a straw-man proposal to make four related policy changes. The proposed changes are outlines below along with a summary of comments received on them. 1. Reduce the minimum allocation size from /20 to /21 - There was some support for this point. There were requests for /21 allocations to come from a new and separately identified address blocks. 2. Remove the requirement to show an immediate need for 25% of the allocated address space (a /23 in this case) - There was no objection to this point. It was pointed out that with a lower barrier to entry the overhead of checking each requester is ready to use 512 IP addresses straight away would be unlikely to be equal to the value of the work. 3. No longer assign PI (Portable) address space to End Users - There some support for to this point. The issue of Root DNS Servers was raised but it was noted that all Root DNS Servers operating in this region already have address assignments. 4. End Users requiring a portable address block could become an LIR and receive a /21 allocation. - There was some support for this point. The costs of operating an LIR were raised as an issue. It was also noted that everyone may become an LIR. There is no barrier to membership of the RIPE NCC. There was a suggestion for a one-time service fee. There were also some additional comments: It was noted that if the policy allows for address allocations based on other criteria than prior demonstrated need some providers may filter those allocations. It was also noted that the RIPE NCC cannot provide any guarantee as to whether address space will or will not be routed of filtered by network operators. Some statistics were requested and provided for the first half of 2003: ASN assignments: 599 Allocation: 377 PI Assignments: 408 Number of /20s allocated per month for the same period: 200301 7 200302 20 200303 14 200304 16 200305 35 200306 24 Finally, it was noted that there is a requirement for globally unique addresses that will not be routed on the Internet.

9 14

Re: [address-policy-wg] Transferring InterNIC space from ARIN Database to other RIR datab ases
by Niall O'Reilly 20 Aug '03

20 Aug '03

On Wednesday, Jul 30, 2003, at 22:42 Europe/Dublin, Hans Petter Holen wrote: > Hmm, we did form a special task force for this matter to work with the > RIPE NCC with the process etc. (References in Leos message) > > As chair of this working group I would really like to know what more we > can do to interact more closesly with the community - please feel free > to > suggest to the list or approach me personally with suggestions ! Hans Petter, Thanks for your encouraging words. I've been away for a couple of weeks and am still mining the e-mail mountain. During second half of July, I've had some off-list exchange with Leo, and found him really eager to be helpful. I still have some sense of the "blind man and elephant" problem, in both directions. As I'm not involved in a LIR, I missed the formation of the task force. In addition, I couldn't travel to Barcelona. In any case, a TF involving LIR-WG and RIPE-NCC is likely (as seems FWIS to be actually the case) to focus on managing the production line rather than on managing customer expectations. It's worth bearing in mind that ERX customers are likely, for the same historical reasons by which they have become ERX customers, to be involved directly with the RIR rather than via their ISP as LIR. [ Hmm, the emergence of the Address Policy WG seems to be just what I needed. ] It seems to me that a useful view of the ERX project and its progress would be a set of graphs derived from a table like the one below; needless to say, my numbers can't be related to reality except by co-incidence. Reporting Period (Month or Qtr) 2003.01 2003.02 ... Estimated work (contact count) 8000 8251 ... (estimate may evolve) Estimated work (network count) 215 216 ... Completed work (contact count) 0 57 ... Completed work (network count) 0 16 ... The graphic would show clearly, for two different measures of the work involved, both the estimated and (cumulative) completed work at each reporting instant. The gap between the estimated and completed work would show how much was left to be done. Hopefully, the curves would converge! I haven't been able to find such a simple overview in the material I've located on the web, or in the nice Member Update which Leo has kindly had sent to me. I think presentation of this kind of information would help customers anticipate ERX progress in the foreseeable, rather than indefinite, future. It would also be useful to know in what order the different /8 blocks are to be dealt with, what criteria have led to choosing this order, and how soon each /8 is expected to be finished, at least as planned at present. I expect that the plan may evolve in the light of experience, and that some blocks of space will prove more troublesome than others. There may be scope for taking advantage of the usual 80/20 result/effort ratio. I expect that the information I would like to see presented must already be in use to manage the project, and so should not require significant extra effort to prepare. It's probably a good idea for me to say that I'm just looking for information, not trying to paint anyone into a corner. [ As you can probably guess, I'm really only interested in 137.43/16, 8-) ] I hope these suggestions help. I expect to come to Amsterdam for the first week in September, and look forward to seeing some nice graphs! Best regards, Niall O'Reilly PS. FWIS: From where I sit.

2 1

Re: [address-policy-wg] IP Addressing policy on personal contact info (kf)
by Michael.Dillon＠radianz.com 20 Aug '03

20 Aug '03

>and I would like to discuss on how we should do with personal contact >information in the RIPE db. The db should only contain the contact information of people who are ready, willing and able to communicate regarding network issues and who are able to act on that communication. This means that much of the existing contact information in the db would disappear and LIRs would formally have the responsibility of relaying communications to their downstream networks. Some of these downstream networks MAY still be in the db but the network contacts MUST agree to be listed and MUST agree to be ready, willing and able to communicate regarding network issue and MUST be able to act on that communication. This would clean most of the useless garbage out of the db, solve the privacy issues, and focus the LIR's attention on the true purpose of the db. The RIPE db is not a general purpose telephone book, it is a contact directory for a specific purpose which is required by TCP/IP internetworking activity. There is no point in including contact data for people who are not ready, willing and able to communicate or who are not able to take any action. RIPE has a direct relationship with LIRs so there is the possibility to force them to have someone who is ready, willing and able to communicate and able to take action. Some LIRs may choose to force their customers to also provide such a contact person, but there is nothing wrong if an LIR chooses to be the direct contact for all of their customers. --Michael Dillon

6 6

Draft agenda Address Policy Working Group RIPE 46
by Hans Petter Holen 20 Aug '03

20 Aug '03

Dear WG; Here are some items for a draft agenda, please send me comments as soon as possible; - wg-charter - policy process - PI policy - initial IPv4 allocation size - RIR- IANA relationship and procedures - RIPE-152 Charging by local registries - Final revised IPv4 policy

5 9

Re: [address-policy-wg] Re: ICANN vs RIPE NCC, was Re: Summary of the PI ......
by Bovio＠aol.com 19 Aug '03

19 Aug '03

In a message dated 11/08/03 17:51:19 W. Europe Daylight Time, daniel.karrenberg(a)ripe.net writes: > I have not stopped listening to concrete ideas about improving the RIPE > NCC. > > I am still here at the RIPE NCC working and listening. Of course the > bureaucratic developments you sneer at *do* happen to some degree. This > is inevitable and both you and I know it. The ability of individuals > like you and me to influence things immediately and directly is reduced. > Of course I personally I do not agree with all things the RIRs do and > more often I do not agree with *how* things are done. > > What seems to divide us is that I still work to improve the 'least of > all evils' structure and you sneer at it providing no alternative. > I would hope that more people will chose the former instead of the latter. > I also hope that people still see the relative mertis and the > differences in legitimacy that exist between the various organisations. > Sneering at the RIPE NCC without suggesting either alternatives or > improvements does not help. > I stand 100% with Daniel here. I can't speak for the others RIRs but I strongly believe the RIPE-NCC has made significant efforts in the recent past to listen to its membership, streamline procedures, and positively react to constructive criticism. There is more work to do, no doubt about it, but I can't see how flaming on mailing lists helps. Whomever has concrete ideas: I propose we move this discussion to the ncc-services-wg list/group, that was created exactly for this purpose. Daniele

8 15

Web archive of PI Task Force now available
by RIPE NCC WebMaster 18 Aug '03

18 Aug '03

Dear Colleagues, As requested, the RIPE NCC has made the discussions of the PI Task Force available as a web archive. You can find the archive at: http://www.ripe.net/ripe/mail-archives/pi-tf/ The PI Task Force was established by the LIR WG at RIPE 44. Its purpose is to review the policy for Provider Independent (PI) address space. Regards, RIPE NCC Webmaster www.ripe.net ==================

1 0

Draft: "status:" re-evaluation
by Shane Kerr 15 Aug '03

15 Aug '03

All, We recently proposed changes to the "status:" attribute values in the RIPE Database: http://www.ripe.net/ripe/draft-documents/status-attributes.html We have received input from APNIC on the subject. They proposed a simpler layout, and we think their suggestion makes sense. However, we would like to go forward with a slightly modified version that incorporates the "organisation" object we are working on. In the meantime, we do not want to delay the implementation of the Sub-Allocation policy. For that reason we'd like to have an interim status attribute value for that. Proposal: Add "SUB-ALLOCATED PA" to the allowed "status:" values. "SUB-ALLOCATED PA" inetnum object may have an "ALLOCATED PA" or an "LIR-PARTITIONED PA" less specific inetnum object. A range of IP's can only have a single "SUB-ALLOCATED PA" in it. That is, you cannot sub-allocate twice. -- Shane Kerr RIPE NCC

6 9

Policy making process
by Hans Petter Holen 15 Aug '03

15 Aug '03

Dear working group, In the beginning there was no documented process of making policy. At some point I started outlining the process in http://www.ripe.net/ripe/wg/lir/howto_develop.html The experience since then is that a couple of elements should be added to this - not necessarily to make a biger or more complicated process, but with the incentive to make the process more open and more transparent. * Proposals should be circulated in writing to the mailinglist in advance of the meetings. Currently we have started calling for contributions and requestion proposals. * There should be detailed technical discussion on the mailinglist before a proposal can be adapted * All policies should be published as RIPE documents * All RIPE documents setting new policies should be published as drafts before becoming final -hph

1 1

IP Addressing policy on personal contact info (kf)
by Katri 14 Aug '03

14 Aug '03

Hello, I saw at RIPE 44 LIR WG Session regarding IP Addressing policy for always-on (residential broadband) Leo said the following: L.V : This should be registered in the database. If it's a residential user, their personal contact information cannot be registered in most countries due to the Data Protection laws. The Database object could then point to contact data at the ISP without identifying customer by name. This should be discussed by the WG if anyone is concerned by this. and I would like to discuss on how we should do with personal contact information in the RIPE db. I don´t only mean residential braodband services but all kind of services that might be available to customers. Should we or should we not register personal contact information that might conflict with Data Protection laws in some countries? How has it been solved earlier? What sort of criterias/ rules should we as an ISP have for the contact data? Regards Katri Forsberg Tele2/Swipnet IP Registry Local Internet Registry __________________________________________________________ Tele2/SWIPnet Tel: 08 5626 40 08 Box 62 Fax: 08 5626 42 10 164 94 Kista Email: ip(a)swip.net SWIP-RIPE SWIPNET-LIR-MNT __________________________________________________________

1 0