[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [dhcwg] Agenda items for dhc WG meeting at IETF67
On Thu, Oct 05, 2006 at 05:47:08AM -0400, Fred Baker wrote:
> Ruri Hiromi presented, during the last IETF, the arguments and
> discussions related to
>
> http://www.ietf.org/internet-drafts/draft-arifumi-ipv6-policy-
> dist-01.txt
> "Practical Usages of Address Selection Policy Distribution", Arifumi
> Matsumoto, 20-Jun-06, <draft-arifumi-ipv6-policy-dist-01.txt>
>
> http://www.ietf.org/internet-drafts/draft-arifumi-v6ops-addr-
> select-ps-00.txt
> "Problem Statement of Default Address Selection in Multi-prefix
> Environment:
> Operational Issues of RFC3484 Default Rules", Arifumi Matsumoto,
> 13-Jun-06,
> <draft-arifumi-v6ops-addr-select-ps-00.txt>
>
> The viewpoint expressed by the working group was that one could
> imagine that a company might have a certain non-default address
> selection policy and that the DHCP option suggested is of value, but
> that the work needed to be done in DHCP WG. What we didn't explicitly
> determine (which the DHCP chairs are asking) is "do people believe
> that companies are likely to build address selection policies that
> materially differ from that of RFC 3484?"
>
> So - what is the opinion of the working group? "Yes, such policies
> can be expected to be sufficiently common that the definition of the
> DHCP option is warranted"? Or "No, such policies are not likely to be
> common"?
One thing we noticed recently was that the newer Fedora install implements
the address selection policy but it now defaults to prefer IPv4 over IPv6.
My understanding of section 2.1 is that IPv6 should, with a default policy,
be preferred over IPv4. It's not clear whether this change in behaviour
was by design or accident.
Thus such an option would be handy either for an admin trying to make sure
that implementations do what the RFC says, or that perhaps want to prefer
IPv4 in their environment, even though both protocols are available.
This bugzilla reference is related and interesting:
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=190495
--
Tim