Re: I-D Action: draft-gont-6man-deprecate-eui64-based-addresses-00.txt

[Alexandru Petrescu <[e-mail]>]

Le 25/10/2013 19:36, Fernando Gont a écrit :
> On 10/25/2013 01:51 PM, Alexandru Petrescu wrote:
>>
>> Not only operationally it may prove difficult to upgrade everything
>> from RFC2464, but all the IPv6-over-foo specs of future deployment
>> may need to accept it as well.
>
> Huh?
>
> Can you elaborate how this affects deployment?

If this spec draft-gont-6man-deprecate-eui64-based-addresses says that
IIDs derived from hardware identifiers are forbidden, and other specs
say that they are mandatory, then there may be a problem.

> Are you suggesting that we ignore well-known problems?

No.

> Are you suggesting that we stick to a feature that allows simple
> location tracking?

Location tracking, as well as other similar characteristics, are bad
when looked at from a prism, and are good when looked at from somewhere
else.

> Or.. maybe you're fine with people deploying IPv6 NAT and that kind
> of middle-box to mitigate this sort of thing?

Yes and no.

IPv6 NAT and NPT are there to be used, and are needed.

No, in that they are needed because more than just its indirect privacy
features.  They are needed also because DHCPv6-PD is not deployed by
cellular operators, and also because SLAAC over Ethernet works with
prefixes length smaller or equal 64.

If you make IIDs which are independent of the hardware ID you'd still
need to make cellular operators deploy DHCPv6-PD and make SLAAC/Ethernet
with longer than 64 prefixes.  Otherwise IPv6 NAT and NPT are still needed.

> An old-timer (who preferred to remain anonymous) just sent me this
> off-list:

Ok.

Alex