Hello all, The ULA discussion has been going on for some time now, and I'd like to summarize it a bit. The differences between ULA address space and PI (or PA) address space: - ULA space should be easier/cheaper to get than PI space - PI space is meant for routing outside your organization and associated networks - ULA space is meant for inside your organization and associated networks Usefulness of ULA space: - I see some people/organizations who would realy like it - I see some people/organizations who don't like it As people who don't ULA don't have to use it (and filter fc00::/7), I would like to see other (preferably objective/technical) reasons why ULA space is a bad idea. Why should we deny ULA space to those who want it and think it is useful to them? Usefulness of ULA-Central space: - Some people think that the possibility of a conflict between two ULA-Local prefixes is so small that it does not really matter - Other people think even that very small chance does matter, and they would like a ULA-Central registry If the people/organizations who want an ULA-Central registry also pay for it, are there any other problems with providing such a registry? The question remains about who should operate and maintain that registry. Because RIPE NCC has a lot of experience with maintaining an IP address registry, they are a likely candidate for this. What arguments are there for and against letting RIPE NCC maintain this registry? What are possible alternatives? I hope I summarized everything correctly, and that I covered all remaining questions. If I missed anything, please let me know. If you have any input about any of the remaining questions, let's discuss it! Thank you, Sander Steffann
Sander Steffann wrote: [..] Please note that the discussions here where about "ULA Central" space, which has a reserved slot at fc00::/8.
As people who don't ULA don't have to use it (and filter fc00::/7),
ULA (RFC4139) is fd00::/8. These are already standardized and accepted and can be automatically generated by anyone. RIR's have no involvement in ULA's at all. Tool for generating ULA's can now be easily found at: http://www.sixxs.net/tools/grh/ula/ Which also contains a 'registry', to 'register' the ULA one is using. This is an intermediate step to ULA-Central. Anybody can generate a prefix and register them in that list. Of course people can simply ignore that list if they want/do not know about it as it is not official in anyway. In that respect, there are some newly started discussions in the IAB about this and on how to move forward with a possible ULA-C scheme. If somebody wants ULA-Central, then, as Fred Baker indicated, use the IETF IPv6 Working Group mailinglist and raise your voice there. When there is again a ULA-Central draft, and then an RFC, then this can be brought forward to the RIR's and am fairly sure that they can then apply a proper policy for it. When there is no such RFC though all of that is useless. My personal stance on all the ULA/PI/PA stuff: RIR's should provide a means to let an organization get a block of size X, where X depends on the size they can justify to the RIR. Greets, Jeroen PS: I heared another rumor that IANA might at one point have a similar "Registered" ULA list, when that happens, the one at SixXS will cease to exist and a redirect will be made to IANA. The list will of course be passed to IANA so that they can build further upon that. The generation function of course will always stay there so that it will be easy to find.
you forgot the DNS part of that... without ULA-C there is no global DNS possibilities. That means split DNS for larger closed network/corporation network. ref RFC4193 - section 4.4... On Wed, 30 May 2007, Sander Steffann wrote:
Hello all,
The ULA discussion has been going on for some time now, and I'd like to summarize it a bit.
The differences between ULA address space and PI (or PA) address space: - ULA space should be easier/cheaper to get than PI space - PI space is meant for routing outside your organization and associated networks - ULA space is meant for inside your organization and associated networks
Usefulness of ULA space: - I see some people/organizations who would realy like it - I see some people/organizations who don't like it
As people who don't ULA don't have to use it (and filter fc00::/7), I would like to see other (preferably objective/technical) reasons why ULA space is a bad idea. Why should we deny ULA space to those who want it and think it is useful to them?
Usefulness of ULA-Central space: - Some people think that the possibility of a conflict between two ULA-Local prefixes is so small that it does not really matter - Other people think even that very small chance does matter, and they would like a ULA-Central registry
If the people/organizations who want an ULA-Central registry also pay for it, are there any other problems with providing such a registry?
The question remains about who should operate and maintain that registry. Because RIPE NCC has a lot of experience with maintaining an IP address registry, they are a likely candidate for this. What arguments are there for and against letting RIPE NCC maintain this registry? What are possible alternatives?
I hope I summarized everything correctly, and that I covered all remaining questions. If I missed anything, please let me know. If you have any input about any of the remaining questions, let's discuss it!
Thank you, Sander Steffann
-- ------------------------------ Roger Jorgensen | - ROJO9-RIPE - RJ85P-NORID roger@jorgensen.no | - IPv6 is The Key! -------------------------------------------------------
Hi,
you forgot the DNS part of that... without ULA-C there is no global DNS possibilities. That means split DNS for larger closed network/corporation network.
ref RFC4193 - section 4.4...
Thanks for pointing that out. Sander
Roger Jorgensen wrote:
you forgot the DNS part of that... without ULA-C there is no global DNS possibilities. That means split DNS for larger closed network/corporation network.
huh? PI space does not support global dns? randy
participants (5)
-
Jeroen Massar
-
Randy Bush
-
Roger Jorgensen
-
Sander Steffann
-
Sander Steffann