Re: [5gangip] Identifier size

Le 01/02/2018 à 18:42, Saleem Bhatti a écrit :
> 
> Tom;
> 
>> On 01 Feb 2018, at 16:50, Tom Herbert <tom@herbertland.com 
>> <mailto:tom@herbertland.com>> wrote:
[...]

>> Use SLAAC to get a /64 to use with application where privacy is not a 
>> concern. Provide DHCPv6 to get single untrackable addresses

If we want the addresses or prefixes to be untrackable then SLAAC may be 
a recommendation, instead of DHCPv6, but.

Much of the trackability in 4G networks comes from the fact that the 
link to UE is a point-to-point link.  There is a single prefix for this 
ptp link, be it with SLAAC or DHCP.

So my question is whether someone works on making the 5G links to the UE 
be more like shared links like in WiFi.  Many UEs already make the link 
to the network look more like a WiFi link than a ptp link: it has a MAC 
address, etc.  But the PGW still considers it as a ptp link.

Alex

  for
>> privacy-sensitive or anonymous communications, Identifier locator 
>> split needs to support both models.
> 
> RFC8064 (PS) / RFC7217 (PS) has your second sentence covered, without 
> needing a DHCPv6 server. (Avoids the DHCPv6 server becoming the tracking 
> device.)
> 
> As far as I can tell, the options for generating the Identifier value 
> are for an ID/LOC system which conforms to RFC8200 (S) for a /64 Locator 
> are:
> 
> - SLAAC: where identity privacy or verifiable identity is not a concern.
> - RFC3972 (PS): where a cryptographically-verifiable identity is required.
> - RFC8064 (PS): where identity privacy is required.
> 
> Possibly, there are other options.
> 
> Cheers,
> --/Saleem
> 
> 
>>
>> Tom
>>
>> _______________________________________________
>> 5gangip mailing list
>> 5gangip@ietf.org <mailto:5gangip@ietf.org>
>> https://www.ietf.org/mailman/listinfo/5gangip
> 
> 
> 
> _______________________________________________
> 5gangip mailing list
> 5gangip@ietf.org
> https://www.ietf.org/mailman/listinfo/5gangip
> 