On 29 Jun 2007, at 7:36pm, Scott Leibrand wrote:
[...]
The RIRs currently require small networks to get provider
aggregatable (PA) space from their provider(s).
This just isn't true. Both APNIC and the RIPE NCC will assign
prefixes longer than an IPv4 /24. APNIC even document the fact quite
nicely where they say "There is no minimum assignment size for
portable assignments made under these terms".
http://www.apnic.net/docs/policy/add-manage-policy.html#11.1
Over the last two years the RIPE NCC has assigned about 2800 prefixes
longer than /22 and APNIC have made over 200. It's all there in the
statistics files that the RIRs publish every day on their FTP sites.
[...]
I think that ULA-C is a really really bad idea because of this.
Perhaps. But just as "Democracy is the worst form of government
there is, except for all the others we've tried", I think ULA-C is
better than what's available now, and more practical in the near
term than a PI-for-everyone utopia.
The fact that IPv6 PI space exists at all is due to the action of
individuals working through the RIR public policy process to create
it. Some of the same individuals, recognizing they've pushed as
far as they can toward liberalizing PI for now, would like to
create additional flexibility for small networks to use IPv6 in a
way that suits their needs. In order to avoid creating an IPv6
"swamp", those individuals have asked the IETF to standardize a new
form of registered private IP space out of a single common global
netblock, and allow the RIRs to begin allocating/assigning that
space to registrants. Their attempts to do so are being met with
cries of "why don't you just liberalize PI." Do you see the irony
here?
I can see that people will grab some ULA-C space if it becomes
available. But I've not yet read a good definition of where it won't
be routable while I've read at least one statement confirming that
ULA-C routes would be accepted if they were offered.
The core problem seems to be a very soft definition of "local". If
that's firmed up then there might be a way to convince people that
ULA-C can be restricted to "local" networks and not be used as a way
of getting cheaper PI space, or a way of getting PI space where the
RIR doesn't have a policy supporting its assignment.
Regards,
Leo
--------------------------------------------------------------------
IETF IPv6 working group mailing list
[email protected]
Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------
