[AusNOG] RFC7217 - "A Method for Generating Semantically Opaque Interface Identifiers with IPv6 Stateless Address Autoconfiguration (SLAAC)."

[Alastair Johnson]

On 5/5/2014 8:12 PM, Mark Newton wrote:
>
> On May 6, 2014, at 12:16 PM, Michael Biber <mbiber at ipv6forum.com.au>
> wrote:
>
>> Thank you Mark. Whenever I do training, workshops or consulting
>> discussions, I conduct a straw poll to see if people do or intend
>> to use SLAAC. The answer is invariably no, with DHCPv6 dominating
>> and static assignment taking up the rest. This varies by
>> role...service provider/end user/DC/content provider.
>
> Service providers shouldn’t be assigning end-user addresses;  they
> should be assigning prefixes.  Then the end user makes the policy
> decision about whether to use SLAAC or not for their hosts.

That implies that an end-user would never attach a host (that only wants 
an address, and quite possibly via SLAAC only) directly to the SP 
service. This, in some topologies, is not the case.

I've watched some ISPs go to quite extensive lengths to make this 
topology work - supporting SLAAC and DHCPv6 on end-user facing 
interfaces so that you can still connect a host directly to the service.