Re: about hardware address

[JINMEI Tatuya / 神明達哉 <Jinmei_Tatuya at isc.org>]

At Fri, 11 Jan 2008 09:52:27 +0900,
Yukiyo Akisada <akisada at tahi.org> wrote:

> I have a question about RFC 4862 Section 5.4.5.
> Please let me confirm about it.
> 
> It says,
> 
>     909    If the address is a link-local address formed from an interface
>     910    identifier based on the hardware address, which is supposed to be
>     911    uniquely assigned (e.g., EUI-64 for an Ethernet interface), IP
>     912    operation on the interface SHOULD be disabled.
> 
> Does the word 'hardware address' here mean MAC address itself?
> Or the address which is assigned physically at the factory?

I'd say it's the latter, according to the background idea of this rule
(which is described in that same section).  Consider the case where
two different hosts located in the same link are manually configured
with the same Ethernet (MAC) address and the same link-local address
is configured for these two interfaces.  Then DAD will detect the
duplication of the link-local address, implicitly revealing the
conflict of the MAC address.  This section intends to detect such a
case and avoid subsequent disruption caused by keeping the conflicted
MAC address.  Whether it's a "real" hardware address or not doesn't
matter in this sense.

So,

> When DAD fails for IP address based on software-based MAC address,
> should the system disable IP operation on also that interface?

the answer to this question is YES.

---
JINMEI, Tatuya
Internet Systems Consortium, Inc.

--------------------------------------------------------------------
IETF IPv6 working group mailing list
ipv6 at ietf.org
Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------