Re: 64bit MAC addresses and SLAAC

[Alexandre Petrescu <[e-mail]>]

Le 16/06/2020 à 17:38, Behcet Sarikaya a écrit :
> 
> 
> On Mon, Jun 15, 2020 at 4:02 PM Bob Hinden <[e-mail] 
> <mailto:[e-mail]>> wrote:
> 
> Alexandre,
> 
>> On Jun 15, 2020, at 1:23 PM, Alexandre Petrescu
> <[e-mail] <mailto:[e-mail]>>
>  wrote:
>> 
>> Hi,
>> 
>> Before the sanitary situation I was studying an issue at ISO.
>> 
>> The issue is about 64bit MAC addresses and SLAAC.
>> 
>> SLAAC needs a 48bit MAC addresses in order to work, and it can
> not work with a 64bit MAC address; (but yes, it can with 64bit 
> IIDs).
> 
> SLACC does not specify the length of the Interface ID, it does not 
> require require 48-bit MAC addresses, and the reason for Modified 
> EUI-64 Format Interface Identifiers in RFC4291 was to support 64bit 
> EUI-64 Identifiers.   We have since moved away from using MAC 
> addresses as Interface IDs.  See RFC 8064.
> 
> 
> 
> +1

Well, I tend to agree with some aspects, but disagree with others.

I agree that putting a MAC address in clear in an IP address raises
significant privacy issues.  These issues should be dealt with, and have
been so.

One way is to use RFC7217 and RFC8064 which, when implemented, offer a
means to put a random number in these many bits of an IID of an IP
address, rather than a hard-wired and thus easily trackable MAC address.

However, that view is not necessarily enough.

There are many problems with RFC7217 plus RFC8064, among which:

- RFC8064 recommends simply stable IIDs; to do that it uses RFC7217
   which defines stables IIDs _within_ a subnet.  That is an
   inconsistency.  A stable IID with a stable prefix is often needed for
   mobility, and, a same IID within distinct prefixes might ease
   administration; yet none of these two aspects seems mentioned.

- neither RFC8064 nor RFC7217 mention the use of random numbers in MAC
   addresses.  Yet that is implemented in Windows at least.

- RFC8064 is somehow not implemented in linux, even though linux does
   offer privacy IIDs.  The part of RFC8064 that is not implemented in
   linux is the part that says that the IID length can be anything and
   not necessarily 64.

Because of these reasons, generally speaking about SLAAC and IID
generation, there might still be other necessary ways to generate IIDs
than just 64bit random IIDs.  Easy to remember IIDs, same IIDs for
distinct interfaces, are a couple of new potential rules that come to mind.

If so, probably the easiness of generating IP addresses containing MAC
addresses might also be a good idea in some context: just copy paste the
64bit (random) MAC address to the last 8 bytes of the IP address.

Alex

> Behcet
> 
> Bob
> 
> 
>> 
>> The document ISO/CD 22738:2019(E) under development at that time
> does specify a 64bit MAC address for optical camera communications 
> (OCC, akin to IEEE 802.15.7).
>> 
>> This might be an additional requirement for SLAAC improvements.
> Or not.
>> 
>> Alex
>> 
>> --------------------------------------------------------------------
>
>>
>> 
> IETF IPv6 working group mailing list
>> [e-mail] <mailto:[e-mail]> Administrative Requests: 
>> https://www.ietf.org/mailman/listinfo/ipv6 
>> --------------------------------------------------------------------
>
>>
>> 
> --------------------------------------------------------------------
>  IETF IPv6 working group mailing list [e-mail] 
> <mailto:[e-mail]> Administrative Requests: 
> https://www.ietf.org/mailman/listinfo/ipv6 
> --------------------------------------------------------------------
>