February 2003 - ipv6-wg - mailman.ripe.net

[ipv6-wg@ripe.net] Reverse delegation for 6bone address space in ip6.arpa
by Andrei Robachevsky 04 Apr '03

04 Apr '03

Dear Colleagues, In accordance with RFC 3152, the administration of the name space within ip6.arpa has been delegated to the RIRs. The 3FFE::/16 address space has not yet been allocated to the RIRs by the IANA. It is the intention that the RIRs administer this space. The community has expressed a need for 3FFE reverse space to be in the ip6.arpa domain. The RIR CEOs sent a message to the IAB requesting a waiver regarding the RFC 3152 requirement and received a positive response. The RIRs are prepared to administer e.f.f.3.ip6.arpa domain in the manner as described below. We plan to implement it in the next few weeks and deliver the solution soon after the RIPE 44 Meeting in January. Regards, Andrei Robachevsky CTO, RIPE NCC Support for reverse delegation for 6bone address space in ip6.arpa DNS tree --------------------------------------------------------------------------- Phase I. Before a final decision is made on future management of the 3FFE address space The delegation for e.f.f.3.ip6.arpa is made by IANA to one of the RIRs. Zone information for this space is produced by copying the existing delegations in the e.f.f.3.ip6.int zone. Additional check is done to ensure that the servers corresponding to the NS RR are also authoritative in .arpa space. Zone information is updated whenever changes are detected in the e.f.f.3.ip6.int zone. This approach has a couple of advantages. It keeps the RIRs out of the 6bone registration process until a final decision is made on future management of the 3FFE address space. It also implies no procedural changes from a user's perspective since changes are made in only one place, while at the same time it moves responsibility for maintaining the e.f.f.3.ip6.arpa zone to the RIRs. This then prepares for a smoother transition to Phase II. The main constraint of this approach is that it forces registrants to use the same name servers for .arpa as for .int. Phase II. Registration of 3FFE space is transferred to the RIRs A shared zone management process will be implemented similar to that currently implemented by the RIRs for the legacy v4 space (ERX project, http://www.ripe.net/db/erx/) That is, holders of 6bone address space will be served for reverse delegation by the respective RIR in their region, and the generation of the e.f.f.3.ip6.arpa zone can be entirely automated using zone merging.

2 2

[ipv6-wg@ripe.net] IXP networks routing
by Nicolas DEFFAYET 14 Mar '03

14 Mar '03

Hello, We hesitate to allow IXP's /48 in our routing table. http://www.ripe.net/ripe/docs/ipv6-policy-ixp.html => "Networks assigned under this policy may not be globally routable." What's the best practice ? Thanks Best Regards, -- Nicolas DEFFAYET, NDSoftware NDSoftware NOC: http://noc.ndsoftwarenet.com/ FNIX6 (French National Internet Exchange IPv6): http://www.fnix6.net/ EuroNOG: http://www.euronog.org/

11 15

[ipv6-wg@ripe.net] Re: Reverse delegation for 6bone address space in ip6.arpa
by Axel Pawlik 20 Feb '03

20 Feb '03

Petr, all, over the last months, the RIRs have developed a reasonably pragmatic procedure to basically mirror the reverse delegation entries under ip6.int into ip6.arpa. Today, we have a working prototype implementing this approach. Unfortunately, our proposal has not been received favourably by the current administrator of the 6bone reverse delegations. To satisfy the needs of our members, however, we are now discussing another way to provide this functionality in a different way, building on the experiences we have made during the Early Registrations Transfer (ERX) Project. We will coordinate this closely with the other RIRs during next week's APNIC meeting; we are obviously also keeping an eye on the 6bone developements as regards the registry transfer and phaseout proposals. Expect a progress report after that. regards, Axel

3 3

[ipv6-wg@ripe.net] network's architecture under IPv6
by BEGIN, Thomas 20 Feb '03

20 Feb '03

Hello to all, I am wondering on IPv6 impacts over a company network's architecture. I mean, a big IPv4 network has to be divided in several VLANs in order to lighten the available bandwidth (sometimes for security reasons but it is not my interest in this case). And this because broadcast messages are received and examined by all network adapter. Thus comes my question: As there is no more broadcast messages but only multicast messages in IPv6, is there still a need to divide a big network in smaller networks (VLANs) ? Thanks Thomas BEGIN

2 1

[ipv6-wg@ripe.net] network's architecture under IPv6
by BEGIN, Thomas 20 Feb '03

20 Feb '03

Hello to all, I am wondering on IPv6 impacts over a company network's architecture. I mean, a big IPv4 network has to be divided in several VLANs in order to lighten the available bandwidth (sometimes for security reasons but it is not my interest in this case). And this because broadcast messages are received and examined by all network adapter. Thus comes my question: As there is no more broadcast messages but only multicast messages in IPv6, is there still a need to divide a big network in smaller networks (VLANs) ? Thanks Thomas BEGIN

1 0

[ipv6-wg@ripe.net] Second IPv6 deployment survey
by leo vegoda 20 Feb '03

20 Feb '03

Dear Colleagues, At the IPv6 WG session at RIPE 44 last month, the RIPE NCC was requested to conduct a second survey on IPv6 deployment. Information on the first IPv6 deployment survey conducted by the RIPE NCC is available at: <http://www.ripe.net/ipv6/ipv6-survey-summary.html> <http://www.ripe.net/ripe/meetings/archive/ripe-42/presentations/ripe 42-ipv6-survey/index.html> In the last survey we contacted LIRs that had received IPv6 allocations. To help develop the structure of the second survey, we would like community input on the following issues: 1. Should the survey be restricted to RIPE NCC members? 2. Should the survey be restricted to those holding IPv6 address space? 3. If the answer to 2 is "yes", should those networks have had the address space for any particular length of time? We would also appreciate comments on issues that are important to the IPv6 community with regards to IPv6 deployment. We look forward to your input. Kind regards, -- leo vegoda RIPE NCC Registration Services

2 1

[ipv6-wg@ripe.net] New IPv6 block allocated to RIPE NCC
by leo vegoda 17 Feb '03

17 Feb '03

Dear Colleagues, The RIPE NCC received the IPv6 address range 2001:1400::/23 from the IANA in February 2003. You may wish to adjust any filters you have in place accordingly. More information on the IP space administered by the RIPE NCC can be found on our web site at: <http://www.ripe.net/ripe/docs/smallest-alloc-sizes.html> Kind regards, -- leo vegoda RIPE NCC Registration Services

2 1

[ipv6-wg@ripe.net] IPv6, future internet, hierarchy
by Andrius Kasparavicius 14 Feb '03

14 Feb '03

Hello, what I should think about when I allocate IPv6? :-) Just like IPv4 subnets for needed IP devices, the same but multipled(how much?) for future hierarchy, divide all I have for all who need(best hierarchy). Divide geographically, by organisations, people population for best hierarchy all v6? I look ahead, hierarchy.. a lot of IPv6. Hmm.. And at the end, would I be bad if I will use /127 for point-to-point(tunnels). respectfully, Andrius

7 29

RE: [ipv6-wg@ripe.net] IPv6, future internet, hierarchy
by Michel Py 14 Feb '03

14 Feb '03

Gert, >> Michel Py wrote: >> available for that. What Gert wants is a large number of >> bits to make the subnet number look good, in the *subnet* >> bits, > Gert Doering wrote: > Call it "look good", but at least you understand the goal. > I appreciate that. Don't get me wrong: the goal (like any naming convention) is good (although operational experience suggests that in this very case it will be useless after a few years). This is not the point. The point is that you think that you can use the IID bits for routing purposes. This discussion has come five times on the table in the last 10 years and has always ended in the hole. In a nutshell, what you want is to declare that the IID bits could have a mixed use, either as IID bits or as subnet bits. Sharing the IID bits for routing purposes is not a good idea, and we just don't have enough bits to do what you want otherwise. > At least four people have voiced here that they are doing this, > and I'm pretty sure that many more networks are doing non-/64s > on point-to-point links for similar or other reasons) As Randy and I said before, the ipv6 WG would be the place to discuss addressing architecture issues. I will clue you though that what some call the "old guard" includes lots of very knowledgeable people with CONSIDERABLE operational experience, including a number of people that have tried what you suggest before with other protocols. Large address spaces such as IPv6 are nothing new to some of us. More than a decade ago, I was dealing with a lot of Novell/IPX networks. The IPX address is 80 bits, 48 IID bits and 32 subnet bits (Hex notation same as IPv6). We did see a lot of funny subnets numbers such as "BABE4B0B" and a few other more explicit ones that I can't really mention here. They were contests to name subnets in the good old IPX days. Anyway, at the time these 32 subnet bits were more than enough to do what you are suggesting for IPv6, some people including myself actually tried, and after reorganizing and merging networks it becomes meaningless. In short: Nice idea, bit a) it does not work in the long run, we tried it before and b) even if it did the benefits that it brings are not worth running the IID bits in mixed mode anyway. Michel.

2 1

RE: [ipv6-wg@ripe.net] IPv6, future internet, hierarchy
by Michel Py 14 Feb '03

14 Feb '03

Stephane, > Stephane Bortzmeyer wrote: > Clearly, it is not so simple. While you cannot encode > *everything* even in 128 bits (billing status), it is > better when you can encode at least a few things. Yes, it is better when you *can*. > Otherwise, "serious" operators would not name their > routers with sensible names like > P12-0.NYKCR2.New-york.opentransit.net This takes 296 bits, btw. It is done because it *can* be done because it's a name, and that's exactly what names have been designed for. Not IP addresses. This is not the issue anyway: We are not talking about configuring the MAC address with a structure that is meaningful; this can be done already and there are 62 bits available for that. What Gert wants is a large number of bits to make the subnet number look good, in the *subnet* bits, totally ignoring conservation. Where does this come from? Nobody does that except him, nobody does it in v4, it has never been part of v6. We can't allow end-sites to do that kind of thing, they would all need at least 48 subnet bits as well. Michel.

3 2