On 2010-09-09 06:18, [email protected] wrote:
>> Now, operators wanted to offer IPv6 service. I hope we think that is a
>> good thing. For residential, they looked at what they could count on
>> from the hosts. And some of them concluded that they could not count on
>> DHCP, so they designed an architecture around SLAAC. In other words,
>> they ddi what we told them to do.
>
> On the other hand, some operators have concluded that RA/SLAAC is simply
> insufficient. Currently, they need to use RA plus some other mechanism
> (for instance DHCP) - but they would *prefer* to use only one mechanism.
>
> Currently, SLAAC/RA alone cannot supply all the necessary parameters
> - but people seem to want to put more and more into RA/SLAAC. That
> would be fine by me *if* we can also have a full featured DHCP which
> can operate without RA. Yes, I know there's strong political/religious
> resistance to doing this - but I believe this is where we're headed...
Firstly, I'm all in favour of fully featured DHCPv6 and I expect that it
will be included in all successful v6 stacks, simply because there will
be sites and providers that require it. I don't think the IETF needs
to worry about that much, though I'd certainly see it as a MUST
implement if I was a vendor.
But there are very strong reasons that are neither political nor
religious for retaining SLAAC as the default mechanism for starting
up a network, and I think that has to include RAs to make any sense
whatever. Basically, we need zeroconf networks to be viable. I can't
see why that would be a problem for an operator who uses DHCPv6 as their
supported mechanism.
I don't see any technical case for expanding SLAAC/RA beyond
RFC 5006, and that's already in hand.
Brian
--------------------------------------------------------------------
IETF IPv6 working group mailing list
[email protected]
Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------
