Hi! The danish research network is part of the native IPv6 project 6net. We've tried for some months to get some IPv6 addresses assigned for this, but the active guidelines - specifically the demand for three peers in the default-free zone, a demand that can't be met for political and economical reasons - made this impossible. I gathered that changes were a-coming and RIPE-42 might be a turning point in this. However, I was unable to attend, so I'd like to hear if any consensus was achieved in this? Peter B. Juul, Uni�C (PBJ255-RIPE)
In your previous mail you wrote: The danish research network is part of the native IPv6 project 6net. => you should try to get a 6bone prefix... Peter B. Juul, Uni�C (PBJ255-RIPE) => BTW Uni-C is one of the three first members of the 6bone. Regards Francis.Dupont@enst-bretagne.fr
The document containing the revised allocation and assignment policy: IPv6 Address Allocation and Assignment Global Policy ftp://ftp.cs.duke.edu/pub/narten/ietf/global-ipv6-assign-2002-04-25.txt was discussed (in the LIR WG meeting) and consensus was achieved within the RIPE community (joining consensus achieved in APNIC's and ARIN's communities).... I suspect the doc will answer your question. philip -- At 13:37 10/05/2002 +0200, Peter B . Juul wrote:
Hi!
The danish research network is part of the native IPv6 project 6net.
We've tried for some months to get some IPv6 addresses assigned for this, but the active guidelines - specifically the demand for three peers in the default-free zone, a demand that can't be met for political and economical reasons - made this impossible.
I gathered that changes were a-coming and RIPE-42 might be a turning point in this.
However, I was unable to attend, so I'd like to hear if any consensus was achieved in this?
Peter B. Juul, Uni�C (PBJ255-RIPE)
On Fri, 10 May 2002, Peter B . Juul wrote:
Hi!
The danish research network is part of the native IPv6 project 6net. We've tried for some months to get some IPv6 addresses assigned for this, but the active guidelines - specifically the demand for three peers in the default-free zone, a demand that can't be met for political and economical reasons - made this impossible. I gathered that changes were a-coming and RIPE-42 might be a turning point in this.
However, I was unable to attend, so I'd like to hear if any consensus was achieved in this?
Hi Peter and all, I'm working for Turkish Academic Network, and wonder what is the main idea for "THE DEMAND for THREE PEERS IN THE DEFAULT-FREE ZONE" ? This is also a demand for us that can't be met for political and economical reasons for now (honestly for 1 years) I also aggre with Peter and suggest attendees of RIPE-42 meeting to discuss about this requirement/demand. Thank you, Yucel Guven Turkish Academic Network-ULAKNET @ AS8517 tel:+90-312-2989311 yucel@ulak.net.tr fax:+90-312-2989393
On Fri, May 10, 2002 at 04:39:33PM -0200, Yucel Guven wrote:
I also aggre with Peter and suggest attendees of RIPE-42 meeting to discuss about this requirement/demand.
Ah, sorry, no, the RIPE-42 ended on may 3rd. I was just asking if this had been discussed, as I'd heard from other sources it would be. The minutes are sometimes a while in the making (sorry, David) so I just wanted to see if someone could give a short reply on that specific point. Peter B. Juul, Uni�C (PBJ255-RIPE)
Hi, On Fri, May 10, 2002 at 03:44:28PM +0200, Peter B . Juul wrote:
On Fri, May 10, 2002 at 04:39:33PM -0200, Yucel Guven wrote:
I also aggre with Peter and suggest attendees of RIPE-42 meeting to discuss about this requirement/demand.
Ah, sorry, no, the RIPE-42 ended on may 3rd. I was just asking if this had been discussed, as I'd heard from other sources it would be. The minutes are sometimes a while in the making (sorry, David) so I just wanted to see if someone could give a short reply on that specific point.
That specific point is history. There was consensus that the outstanding new "Global IPv6 interim policy draft" (that was posted just a few weeks ago in a revised edition) was to be made new policy. Wether that helps you depends on your network - criteria for getting IPv6 allocations boil down to "you must be willing to assign IPv6 networks to a substantial number of third party sites". If you're assigning to research institutes, universities, and so on, the criteria should be easily met. If you want addresses only for your own transport/server infrastructure, the new policy will not get you public space for that. You could use upstream space (recommended), or 6bone space (maybe not so good an idea). Gert Doering -- NetMaster -- Total number of prefixes smaller than registry allocations: 45077 (47584) SpaceNet AG Mail: netmaster@Space.Net Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0 80807 Muenchen Fax : +49-89-32356-299
On Fri, 10 May 2002, Gert Doering wrote:
On Fri, May 10, 2002 at 03:44:28PM +0200, Peter B . Juul wrote:
On Fri, May 10, 2002 at 04:39:33PM -0200, Yucel Guven wrote:
I also aggre with Peter and suggest attendees of RIPE-42 meeting to discuss about this requirement/demand.
Ah, sorry, no, the RIPE-42 ended on may 3rd. I was just asking if this had been discussed, as I'd heard from other sources it would be. The minutes are sometimes a while in the making (sorry, David) so I just wanted to see if someone could give a short reply on that specific point.
That specific point is history. There was consensus that the outstanding new "Global IPv6 interim policy draft" (that was posted just a few weeks ago in a revised edition) was to be made new policy.
Wether that helps you depends on your network - criteria for getting IPv6 allocations boil down to "you must be willing to assign IPv6 networks to a substantial number of third party sites". If you're assigning to research institutes, universities, and so on, the criteria should be easily met.
(The last paragraph: ) Wrong. Section d), assingning at least 200 /48's is new compared to previous RIPE-only draft policy. This is a requirement that most NREN's can *not* honestly meet. -- Pekka Savola "Tell me of difficulties surmounted, Netcore Oy not those you stumble over and fall" Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
On Fri, 10 May 2002, Pekka Savola wrote:
(The last paragraph: ) Wrong.
Section d), assingning at least 200 /48's is new compared to previous RIPE-only draft policy.
This is a requirement that most NREN's can *not* honestly meet.
Amusing that the very people leading IPv6 deployment in Europe may well have to do so with 6bone address space, if the current policy is adopted. Luckily most of the NRENs already have production RIPE allocations, but I guess the rules for NRENs are rather different as the customers are universities not home users (although universities may offer IPv6 dialup or ADSL to its end users so needs its own /48's, but that's another story). Tim
Hi, On Fri, May 10, 2002 at 06:04:19PM +0100, Tim Chown wrote:
(The last paragraph: ) Wrong.
Section d), assingning at least 200 /48's is new compared to previous RIPE-only draft policy.
This is a requirement that most NREN's can *not* honestly meet.
Amusing that the very people leading IPv6 deployment in Europe may well have to do so with 6bone address space, if the current policy is adopted.
The current policy *has been already* adopted. There is no "if" here. Nevertheless I don't buy this "most NRENs can not hoestly meet" paragraph - and again: there have been enough NREN representatives at the RIPE meeting, and I can't remember hearing loud arguments from them against the new policy. There have been questions, and concerns, but most of them have been convinced that the new policy is an improvement, and anybody serious about IPv6 *distribution* should be able to work with it, while others (non-distributing -> end sites, de facto) should be able to get IPv6 addresses as well today. Gert Doering -- NetMaster -- Total number of prefixes smaller than registry allocations: 45114 (45077) SpaceNet AG Mail: netmaster@Space.Net Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0 80807 Muenchen Fax : +49-89-32356-299
- and again: there have been enough NREN representatives at the RIPE meeting, and I can't remember hearing loud arguments from them against the new policy. There have been questions, and concerns, but most of them have been convinced that the new policy is an improvement, [snip]
For the record, I raised a point at the lir-wg, on behalf of HEAnet and something that has met with agreement from other NREN personnel. The requirement for 200 sites, regardless of density to the right of the /48 boundary, is very onerous. However from other discussions it appears that the consequences of rejecting the policy would have been truly dire for some in the APNIC region, so no one wanted to derail the process just for this; instead I think it was recorded as approved, with the acknowledgement that further work needs to be done. Dave -- dave.wilson@heanet.ie ------- DW238-RIPE ------- GPG key: davew+pgp@heanet.ie HEAnet Limited, Brooklawn House, Crampton Ave, Ballsbridge, D4 p:353-1-6609040 "A reload a day keeps the TAC away" - Roger Gottsponer RIPE-40 f:353-1-6603666 HEAnet's National Networking Conference http://www.heanet.ie/conferences/2001/
Hi, On Fri, May 10, 2002 at 07:22:09PM +0100, Dave Wilson wrote:
- and again: there have been enough NREN representatives at the RIPE meeting, and I can't remember hearing loud arguments from them against the new policy. There have been questions, and concerns, but most of them have been convinced that the new policy is an improvement, [snip]
For the record, I raised a point at the lir-wg, on behalf of HEAnet and something that has met with agreement from other NREN personnel. The requirement for 200 sites, regardless of density to the right of the /48 boundary, is very onerous.
And if I remember correctly, it was voiced clearly that the "200" is by no means a hard figure - if someone demonstrates the clear wish (and process) in assigning to "more than a hand ful" of different organizations, it is not meant that the allocation should be taken back again. I still haven't heard why this should be so difficult for an NREN. Just assign /48s to sufficiently small "end sites". Define "end site" according to your needs - there is no formal official definition (except maybe "a /48 only to those that need more than one subnet").
However from other discussions it appears that the consequences of rejecting the policy would have been truly dire for some in the APNIC region, so no one wanted to derail the process just for this; instead I think it was recorded as approved, with the acknowledgement that further work needs to be done.
Especially as the old policy was no better for anybody. It has been clearly said that this is an *interim* policy. So feel free to suggest better criteria to achieve: - everybody that wants to connect a substantive amount of other entities ("end sites") to the IPv6 world can get an allocation - end sites will NOT get an allocation (there is a strong voice from the ARIN region and some agreement from the APNIC people on that point) If we can find something that will fulfill both criteria but not have a "200" in it, I'll be happy to see it thrown out, but nevertheless I think the current policy is a big improvement and should be fine for anybody (except end sites that want to do BGP multihome and their own allocation - but yes, those aren't meant to get one). Gert Doering -- NetMaster -- Total number of prefixes smaller than registry allocations: 45114 (45077) SpaceNet AG Mail: netmaster@Space.Net Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0 80807 Muenchen Fax : +49-89-32356-299
On Fri, May 10, 2002 at 07:22:09PM +0100, Dave Wilson wrote:
- and again: there have been enough NREN representatives at the RIPE meeting, and I can't remember hearing loud arguments from them against the new policy. There have been questions, and concerns, but most of them have been convinced that the new policy is an improvement, [snip]
For the record, I raised a point at the lir-wg, on behalf of HEAnet and something that has met with agreement from other NREN personnel. The requirement for 200 sites, regardless of density to the right of the /48 boundary, is very onerous.
However from other discussions it appears that the consequences of rejecting the policy would have been truly dire for some in the APNIC region, so no one wanted to derail the process just for this; instead I think it was recorded as approved, with the acknowledgement that further work needs to be done.
Yes, this is a good summary. The new policy was adopted, and it was noted that there could be some deficiencies for for example NRENs, large multinational companies or mixed companies that are both in the business of being an ISP and some other line of business. However, it was also mentioned during the meeting, and now again on this mailing list, that the requirements might not be so difficult at all for such organizations. It is probably useful for the earlier mentioned organizations to read the actual policy document, decide whether they actually should go to an upstream or that there is indeed a way for them to qualify and only as the *last* option to raise an issue on the list (that is the lir-wg list, not this list) *after* their application has been rejected and no way out, while working with the RIPE NCC, was found. It's simply a bit premature to say that something doesn't work for your organization if people obviously have not read the policy document, before your organization has applied for address space and before trying to work with the RIPE NCC to work out any issues with the application. David K. ---
On Fri, May 10, 2002 at 12:02:27PM -0700, David Kessens wrote:
However, it was also mentioned during the meeting, and now again on this mailing list, that the requirements might not be so difficult at all for such organizations. It is probably useful for the earlier mentioned organizations to read the actual policy document, decide whether they actually should go to an upstream or that there is indeed a way for them to qualify and only as the *last* option to raise an issue on the list (that is the lir-wg list, not this list) *after* their application has been rejected and no way out, while working with the RIPE NCC, was found.
I don't mean to nag, but the "upstream" situation was exactly why the former policy was a major problem for me. My only upstream provider is NORDUnet, which is a coalition of NRENs in the nordic area. NORDUnet is a transit provider only (from a routing POV) and have no aggregations from which to assign addresses. As far as I know NORDUnet gets some addresses for services from Sunet, but that's all. Thus there was no upstream provider to go to for addresses. (However, I am confident that things are going to work out all right with the new policy.) Peter B. Juul, Uni�C (PBJ255-RIPE)
Tim Chown <tjc@ecs.soton.ac.uk> writes:
On Fri, 10 May 2002, Pekka Savola wrote:
(The last paragraph: ) Wrong.
Section d), assingning at least 200 /48's is new compared to previous RIPE-only draft policy.
This is a requirement that most NREN's can *not* honestly meet.
Amusing that the very people leading IPv6 deployment in Europe may well have to do so with 6bone address space, if the current policy is adopted.
How difficult is it for a NREN to produce 200 valid assignments? Take 200 students who want to participate in IPv6 and potentially have more than one network, and you're there. The assignments might not be *needed*, and they might be done for the sole reason of getting the allocation, but it's certainly in accordance to the guidelines, and thus you do meet the guidelines. Thus you fulfil both the spirit and the wording of the guidelines. I don't see a problem with this. Robert
On Fri, May 10, 2002 at 07:29:25PM +0000, Robert Kiessling wrote:
How difficult is it for a NREN to produce 200 valid assignments? Take 200 students who want to participate in IPv6 and potentially have more than one network, and you're there. The assignments might not be *needed*, and they might be done for the sole reason of getting the allocation, but it's certainly in accordance to the guidelines, and thus you do meet the guidelines.
Thus you fulfil both the spirit and the wording of the guidelines. I don't see a problem with this.
The new policy shouldn't be seen as the "policy to end all policies". It should be seen - I believe - as an interim reply to APNICs "cry for help" at RIPE-40. I believe it's more important that we do this right than that we follow the exact wording of an interim policy. If we take into account Gert's very sensible comments, I don't think a serious provider of IPv6 service is going to get any real problems. Peter B. Juul, Uni�C (PBJ255-RIPE)
On 10 May 2002, Robert Kiessling wrote:
How difficult is it for a NREN to produce 200 valid assignments? Take 200 students who want to participate in IPv6 and potentially have more than one network, and you're there. The assignments might not be *needed*, and they might be done for the sole reason of getting the allocation, but it's certainly in accordance to the guidelines, and thus you do meet the guidelines.
As Pekka points out, this isn't desirable or honest (though JANET will soon have an ADSL service, it is IPv4-only). Tim
Hi, On Fri, May 10, 2002 at 07:42:07PM +0300, Pekka Savola wrote:
Wether that helps you depends on your network - criteria for getting IPv6 allocations boil down to "you must be willing to assign IPv6 networks to a substantial number of third party sites". If you're assigning to research institutes, universities, and so on, the criteria should be easily met.
(The last paragraph: ) Wrong.
Section d), assingning at least 200 /48's is new compared to previous RIPE-only draft policy.
This is a requirement that most NREN's can *not* honestly meet.
Why? I am not a NREN, so I have to rely on what other people tell me, and Wilfried did sound pretty convincing when he claimed that it is easily met if you define "site" accordingly. After all, every department or even every student @ home can be a site. Gert Doering -- NetMaster -- Total number of prefixes smaller than registry allocations: 45114 (45077) SpaceNet AG Mail: netmaster@Space.Net Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0 80807 Muenchen Fax : +49-89-32356-299
On Fri, 10 May 2002, Gert Doering wrote:
On Fri, May 10, 2002 at 07:42:07PM +0300, Pekka Savola wrote:
Wether that helps you depends on your network - criteria for getting IPv6 allocations boil down to "you must be willing to assign IPv6 networks to a substantial number of third party sites". If you're assigning to research institutes, universities, and so on, the criteria should be easily met.
(The last paragraph: ) Wrong.
Section d), assingning at least 200 /48's is new compared to previous RIPE-only draft policy.
This is a requirement that most NREN's can *not* honestly meet.
Why? I am not a NREN, so I have to rely on what other people tell me, and Wilfried did sound pretty convincing when he claimed that it is easily met if you define "site" accordingly.
After all, every department or even every student @ home can be a site.
Yes. It would be an act of extreme stupidity (IMO) if NREN's started giving addresses and *providing connectivity* (directly) as required by the policy for every department, student @ home, etc. (This must be delegated to universities etc. for practical reasons.) This is against the spirit of "/48 to every organization, no matter how big". This is why is said _honestly_. :-) So let's see: an university of less than about 65,000 people will need a /28 to cope with this (with HD-ratio 80%). Assume a NREN has 20 (usually a lot more but some are small) of these. NREN then requires at least /22 (with HD ratio 80%) to cope with these allocations. Do we really want to go down this road? -- Pekka Savola "Tell me of difficulties surmounted, Netcore Oy not those you stumble over and fall" Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
Hi, On Sat, May 11, 2002 at 09:56:25AM +0300, Pekka Savola wrote:
It would be an act of extreme stupidity (IMO) if NREN's started giving addresses and *providing connectivity* (directly) as required by the policy for every department, student @ home, etc. (This must be delegated to universities etc. for practical reasons.)
There's nothing wrong with hierarchy :-) - so the NREN can route a /42 (for example) to a university, out of which individual /48s are assigned to the universities departments. Of course the NREN would have to have the last word in assignment of the /48s, but that doesn't mean the /48 have to be connected directly to a NREN router. (This is why we use HD ratio - because if you do hierarchy, you will never achieve maximum address space usage - and the HD ratio rule takes this into account)
This is against the spirit of "/48 to every organization, no matter how big".
Not really. It's also "/48 to every organization, no matter how *small* (if it is subnetting)". It depends very much of what makes sens as an "organization" in your context. [..]
So let's see: an university of less than about 65,000 people will need a /28 to cope with this (with HD-ratio 80%). Assume a NREN has 20 (usually a lot more but some are small) of these. NREN then requires at least /22 (with HD ratio 80%) to cope with these allocations.
Do we really want to go down this road?
No, but nobody said you have to assign a /48 to every single person working at the university. A useful thing might be to assign /48s to every campus, or every larger department, and maybe one /48 per student hostel (hosting "many" students, each of them gets a /64). So with some reason instead of a "this can't work!" attitude, I think this can work well - if one insists on doing non-useful things, it will break (but yes, this is a problem with the "one site" = /48 rule, because it's too vague to work unless people are reasonable). Gert Doering -- NetMaster -- Total number of prefixes smaller than registry allocations: 45114 (45077) SpaceNet AG Mail: netmaster@Space.Net Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0 80807 Muenchen Fax : +49-89-32356-299
On Sat, 11 May 2002, Gert Doering wrote:
On Sat, May 11, 2002 at 09:56:25AM +0300, Pekka Savola wrote:
It would be an act of extreme stupidity (IMO) if NREN's started giving addresses and *providing connectivity* (directly) as required by the policy for every department, student @ home, etc. (This must be delegated to universities etc. for practical reasons.)
There's nothing wrong with hierarchy :-) - so the NREN can route a /42 (for example) to a university, out of which individual /48s are assigned to the universities departments.
Giving out /48's for departments is wrong. They are not separate organizations.
This is against the spirit of "/48 to every organization, no matter how big".
Not really. It's also "/48 to every organization, no matter how *small* (if it is subnetting)".
It depends very much of what makes sens as an "organization" in your context.
Departments etc. are not separate entities. I don't think university-like organizations ever need more than a /48. One and the only really problematic thing is if they provide access to students/staff/etc., e.g. via DSL, dial-up, dorms or what. /64 would usually be ok (except very large universities and the like), /48 would not. This is why I feel that dial-up/etc. assignments, if done using /48, should be delegated from separate, earmarked address blocks.
[..]
So let's see: an university of less than about 65,000 people will need a /28 to cope with this (with HD-ratio 80%). Assume a NREN has 20 (usually a lot more but some are small) of these. NREN then requires at least /22 (with HD ratio 80%) to cope with these allocations.
Do we really want to go down this road?
No, but nobody said you have to assign a /48 to every single person working at the university. A useful thing might be to assign /48s to every campus, or every larger department, and maybe one /48 per student hostel (hosting "many" students, each of them gets a /64).
IAB/IESG note recommends /48 for dial-up's, DSL's etc. Sometimes universities etc. want to provide access for people. So this is IMO a reasonable assumption (if there were no addressing constraints).
So with some reason instead of a "this can't work!" attitude, I think this can work well - if one insists on doing non-useful things, it will break (but yes, this is a problem with the "one site" = /48 rule, because it's too vague to work unless people are reasonable).
As I've been saying for some time now, the dial-ups, home users etc. are really the only problematic thing here. How do you think that e.g. Nokia, Cisco or Microsoft will do this? I think everyone agrees that /48 should be very well enough, except for the possibility of dial-ups, DSL, remote access etc. -- Pekka Savola "Tell me of difficulties surmounted, Netcore Oy not those you stumble over and fall" Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
On Fri, May 10, 2002 at 04:08:21PM +0200, Gert Doering wrote:
Wether that helps you depends on your network - criteria for getting IPv6 allocations boil down to "you must be willing to assign IPv6 networks to a substantial number of third party sites". If you're assigning to research institutes, universities, and so on, the criteria should be easily met.
Goody goody. We are an ISP for about 200 institutions, several of which are universities who again provide for a large number of institutes. We will of course provide IPv6 for any institution interested. This should solve our problem. Peter B. Juul, Uni�C (PBJ255-RIPE)
Hi, On Fri, May 10, 2002 at 08:35:55PM +0200, Peter B . Juul wrote:
On Fri, May 10, 2002 at 04:08:21PM +0200, Gert Doering wrote:
Wether that helps you depends on your network - criteria for getting IPv6 allocations boil down to "you must be willing to assign IPv6 networks to a substantial number of third party sites". If you're assigning to research institutes, universities, and so on, the criteria should be easily met.
Goody goody. We are an ISP for about 200 institutions, several of which are universities who again provide for a large number of institutes. We will of course provide IPv6 for any institution interested. This should solve our problem.
That's what I meant - if you're serious about IPv6, fulfilling those criteria is really doable. I want to repeat this (for the records): even if the policy talks about "plan to connect 200 sites in the next two years", it does NOT mean "if you only manage to get 190, RIPE will take your allocation away!". It's mean that you - have plans to "build something real" - demonstrate (by documenting) that you're really working on it - that means "not register 20 /48s immediately and nothing ever after" And if after 2 years you can show that you have been growing something like "slowly but steady" and will 200 "eventually", this is expected to be *fine*. Gert Doering -- NetMaster -- Total number of prefixes smaller than registry allocations: 45114 (45077) SpaceNet AG Mail: netmaster@Space.Net Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0 80807 Muenchen Fax : +49-89-32356-299
On Fri, 10 May 2002, Gert Doering wrote:
It's mean that you
- have plans to "build something real"
- demonstrate (by documenting) that you're really working on it - that means "not register 20 /48s immediately and nothing ever after"
And if after 2 years you can show that you have been growing something like "slowly but steady" and will 200 "eventually", this is expected to be *fine*.
But is that what the policy says? :-) Tim
Hi, On Sat, May 11, 2002 at 09:09:37PM +0100, Tim Chown wrote:
It's mean that you
- have plans to "build something real"
- demonstrate (by documenting) that you're really working on it - that means "not register 20 /48s immediately and nothing ever after"
And if after 2 years you can show that you have been growing something like "slowly but steady" and will 200 "eventually", this is expected to be *fine*.
But is that what the policy says? :-)
Not in the very same words, but this is the interpretation that was presented at RIPE42, and the consensus to accept the policy is based on that. So I am going to trust the RIPE hostmasters to get this right (otherwise I'll come and discuss 'em to death... :-) ). Gert Doering -- NetMaster -- Total number of prefixes smaller than registry allocations: 45114 (45077) SpaceNet AG Mail: netmaster@Space.Net Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0 80807 Muenchen Fax : +49-89-32356-299
In your previous mail you wrote: I'm working for Turkish Academic Network, and wonder what is the main idea for "THE DEMAND for THREE PEERS IN THE DEFAULT-FREE ZONE" ? => sub-TLAs are for ISPs, not for NRENs. The solution for a NREN is either to disguise itself as an ISP, or to join the 6bone. This is also a demand for us that can't be met for political and economical reasons for now (honestly for 1 years) => so you should join the 6bone. I also aggre with Peter and suggest attendees of RIPE-42 meeting to discuss about this requirement/demand. => there was no raised problem because RIPE attendees are mainly from ISPs. Regards Francis.Dupont@enst-bretagne.fr
hi, On Fri, May 10, 2002 at 04:24:06PM +0200, Francis Dupont wrote:
I'm working for Turkish Academic Network, and wonder what is the main idea for "THE DEMAND for THREE PEERS IN THE DEFAULT-FREE ZONE" ?
=> sub-TLAs are for ISPs, not for NRENs. The solution for a NREN is either to disguise itself as an ISP, or to join the 6bone.
Well - many NRENs do ISP-like work for their members (universities, student hostels, etc.), so you don't have to "disguise" yourself. If you assign /48s to third parties, you fulfill the criteria - you don't have to "be an ISP" for that. [..]
I also aggre with Peter and suggest attendees of RIPE-42 meeting to discuss about this requirement/demand. => there was no raised problem because RIPE attendees are mainly from ISPs.
I saw a fair number of NREN people at RIPE42, and they did not raise any major concerns... "consensus building" doesn't mean "the majority decides" - it's better than democracy .-) Gert Doering -- NetMaster -- Total number of prefixes smaller than registry allocations: 45077 (47584) SpaceNet AG Mail: netmaster@Space.Net Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0 80807 Muenchen Fax : +49-89-32356-299
On Fri, May 10, 2002 at 04:27:57PM +0200, Gert Doering wrote:
Well - many NRENs do ISP-like work for their members (universities, student hostels, etc.), so you don't have to "disguise" yourself.
In all reasonable definitions of "ISP" we are one - we provide all sorts of Internet services (access, DNS, mail, web-hosting and so on), we peer on local IX'es (actually the same people that run the NREN operate the DIX) etc., so I don't think that will be a problem.
"consensus building" doesn't mean "the majority decides" - it's better than democracy .-)
It is, as long as a consensus can actually be reached :-) Peter B. Juul, Uni�C (PBJ255-RIPE)
Yucel, On Fri, May 10, 2002 at 04:39:33PM -0200, Yucel Guven wrote:
I'm working for Turkish Academic Network, and wonder what is the main idea for "THE DEMAND for THREE PEERS IN THE DEFAULT-FREE ZONE" ? This is also a demand for us that can't be met for political and economical reasons for now (honestly for 1 years)
I also aggre with Peter and suggest attendees of RIPE-42 meeting to discuss about this requirement/demand.
Thank you,
Yucel Guven Turkish Academic Network-ULAKNET @ AS8517 tel:+90-312-2989311
^^^^^^ How did you justify your AS# to the RIPE NCC if you have 'political and economical' reasons that don't allow you to peer with at least three peers in the default-free zone (at this point ipv6 peerings tend to cost nothing by the way) ?!? We care about technical arguments here, not political problems. David K. speaking for myself ---
On Fri, May 10, 2002 at 11:35:56AM -0700, David Kessens wrote:
How did you justify your AS# to the RIPE NCC if you have 'political and economical' reasons that don't allow you to peer with at least three peers in the default-free zone (at this point ipv6 peerings tend to cost nothing by the way) ?!? We care about technical arguments here, not political problems.
There are some points to be made around this, but I don't think the ipv6-wg is the right forum (seeing as the actual demand in RIPE-195 is for IPv4 peers, which is also the case for AS#s). I'm not as savvy in the workings of the wgs as I wish I was. Which one would be right? routing-wg? Peter B. Juul, Uni�C (PBJ255-RIPE)
participants (10)
-
Dave Wilson -
David Kessens -
Francis Dupont -
Gert Doering -
Pekka Savola -
Peter B . Juul -
Philip Smith -
Robert Kiessling -
Tim Chown -
Yucel Guven