Re: A common problem with SLAAC in "renumbering" scenarios

[Fernando Gont <[e-mail]>]

Bert,

On 14/2/19 16:30, Manfredi (US), Albert E wrote:
> -----Original Message-----
> From: ipv6 <[e-mail]> On Behalf Of JORDI PALET MARTINEZ
> 
>> I know that IP addresses are considered personal data, but that doesn't mean they must be non-persistent. Only means that third parties (not the ISP) can't store that data without customer consent, and that the ISP can't provide that data to third parties. This is not different if is a non-persistent prefix!
> 
> I've been diligently following this (enlightening) thread. But the privacy aspects of this desire to retain a persistent IPv6 prefix are undeniable. I'm not sure how to reconcile:
> 
> 1. The strong recommendation against persistent or EUI-based IIDs,
> 
> 2. The strong desire to have 64-bit IIDs, and
> 
> 3. This idea of retaining a persistent CPE prefix.
> 
> If the prefix is known to be 64 bits wide, and is persistent, that prefix identifies a household every bit as positively as a persistent 64-bit IID identifies a device. How come we aren't seeing recommendations AGAINST persistent prefixes for CPE devices?

You nailed it ;-)

Yes, there are clear privacy implications to stable prefixes. There has
been some discussion on the topic (we did a bit in RFC7721). See e.g.:
draft-herbert-ipv6-prefix-address-privacy

This is yet another reason for *not* ruling out dynamic prefixes -- if
noting that "brittleness resulting from assuming state in the network is
a bad" is not enough.

Thanks,
-- 
Fernando Gont
SI6 Networks
e-mail: [e-mail]
PGP Fingerprint: 6666 31C6 D484 63B2 8FB1 E3C4 AE25 0D55 1D4E 7492