Re: [dhcwg] Comments - draft-ietf-dhc-l2ra-extensions-00

On Thu, Oct 23, 2008 at 09:40:08AM +0530, Bharat Joshi wrote:
> Apart from this, as mentioned above, that the client's MAC address may be all zeros. [We found a case where a client may set the chaddr to all 0s]

Not a DHCPINFORM by chance, was it?

> Does DHCP server allocates an IP address to such a request?

Well, in theory it's not supposed to happen, and there seem to be
some RFC's floating around for doing DHCP over specific kinds of
physical networks which have no MAC address (or an address that is
too long to fit in the chaddr field).

I think a server could properly allocate an address to a chaddrless
client if a client-identifier had been sent.

But depending on the kind of network they're attached to, this
breaks things like relay forwarding by unicast, so it would be a
pretty weird, crafted situation.

> Ok. So this is what has been proposed in Layer 2 Relay Agent extension draft that L3 RA parse the sub-option and unicast the reply back to the hardware address.

That's a fine plan, with the only proviso being you can't count on
support for the new option; so you still have to be prepared to
receive packets "the old way" until compatible relays deploy.

-- 
Ash bugud-gul durbatuluk agh burzum-ishi krimpatul.
Why settle for the lesser evil?	 https://secure.isc.org/store/t-shirt/
-- 
David W. Hankins	"If you don't do it right the first time,
Software Engineer		     you'll just have to do it again."
Internet Systems Consortium, Inc.		-- Jack T. Hankins