On Thu, 27 Jan 2005, Nick 'Sharkey' Moore wrote:Ethernet-derived addresses are indeed also an issue, but they're hypothetically unique ... so we're back to estimating the inestimable ... are they less likely to collide than 3041 because of this supposed uniqueness, or more likely to collide because of the possibility of human error?
Good question.. The reason why Ethernet MAC addresses would not be unique would probably be either human error (manually configuring the mac address) or a manufacturing error -- and in either case, it's not certain how much analysis on 2^48 would help..
Or, maybe, privacy. If we question the privacy aspects of sending unique identifiers in IPv6 addresses, then we might also question the privacy impact of broadcasting unique 48 bit identifiers in Ethernet or WIFI frames. Rhetorical question: how long will it take before computers systematically configure a new MAC address each time they reboot, or each time they roam to a new network?
Actually, a friend of mine wrote this kind of code (Ethernet address randomization at bootup) to Linux as an alternative to RFC3041 maybe two years ago. I didn't push for it too much because it would have caused confusion under those cases where you have a stationary node and MAC address "locking" is used in Ethernet switches. (And there's AFAICS no good way to automatically determine this..)
But for example on laptops which are expected to roam a bit more, this might make some amount of sense. It might be better than RFC3041 from the L3 perspective, at least.
But the implications should IMHO be explored at some length first (maybe write an I-D?). Changing MAC addresses is going to cause more disruption on the roaming host in case of false positives, i.e., the laptop thinks it moved but actually didn't, because the communication using the old MAC address and resulting IP addresses is going to be disrupted unless you assume some kind of Neighbor Advertisement message use ("gratuitous ARP").
-- Pekka Savola "You each name yourselves king, yet the Netcore Oy kingdom bleeds." Systems. Networks. Security. -- George R.R. Martin: A Clash of Kings
-------------------------------------------------------------------- IETF IPv6 working group mailing list [email protected] Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
