On 18/12/2012 14:13, Fernando Gont wrote:
> On 12/18/2012 10:55 AM, Ole Troan wrote:
>>>>> Nobody even suggested that. For instance, if these addresses had a
>>>>> lifetime (in the RFC4941 sense), they wouldn't be called "stable" in the
>>>>> first place.
>>>> I suggest that you add a discussion of site renumbering considerations.
>>>> The problems described in draft-ietf-6renum-static-problem need
>>>> to be avoided.
>>> Could you please elaborate what you have in mind? (i.e., how you think
>>> this should be addressed, without rehashing the whole discussion in the
>>> aforementioned I-D -- I guess one or two paragraphs, and a reference to
>>> draft-ietf-6renum-static-problem?)
>> my understanding was that addresses based on these interface-ids would have
>> the same address lifetime properties as RFC4861 addresses.
>
> Agreed. What I meant in my response to Hosnieh is that these addresses
> behave (in terms of lifetimes) in the same way as traditional slaac
> addresses, but do not vary over time as RFC4941-addresses.
>
> So it's not clear to me what's the concern here.
The concern is that we are still mainly ignoring the renumbering
problem, and in some years time this will be serious for users.
But if you add "From the point of view of renumbering, these addresses
behave like RFC4941 addresses" the point is covered.
>
>> is it the draft's reference to "static addresses" that creates the confusion?
>
> I had the impression that Brian suggested to include something along the
> lines of "while having table addresses is nice, beware that you should
> probably not rely strongly on them being stable, since a renumbering
> event could make them change" (e.g., DNS is always valuable to contact
> servers, even if the addresses are stable, etc.). Maybe he can
> elaborate.... :-)
I think the above is enough.
Brian
>
> Cheers,
--------------------------------------------------------------------
IETF IPv6 working group mailing list
[email protected]
Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------
