subenet-mask option in DHCPACK for DHCPINFORM

[David W. Hankins]

On Mon, Jul 23, 2007 at 01:37:22PM -0700, Jane Du (jadu) wrote:
> Actually, the dhcp client we are using doesn't like that the subnet-mask
> option is included in the DHCPACK when it doesn't request it. The vendor
> of the client told us that we need to change the code by not sending the
> option because it doesn't follow the spec. In RFC 2131, the option 1 is
> not a "must" added option in the ACK, right? 

It is neither a MUST nor is it a MUST NOT.  In fact, servers MAY
provide additional options.

In the context of DHCPINFORM, I think this behaviour is not right,
and we'll fix this in a future release.

In the case of DHCPREQUEST, this behaviour is unfortunately mandatory
to be compatible (and to actually work on the modern Internet).

I'm a little bit disappointed to see a client that throws an error if
it sees an option it doesn't understand.  The best thing to do is to
just ignore the option.  These sorts of things do happen!

> Do you see any issue if I just change code not to send option 1 when it
> is not requested?

If you don't have any clients with the aforementioned flaw, nothing
will go wrong (for example, if you were working fine with 3.0.x).

> In which version, does ISC server start sending the subnet-mask for the
> ack to request?  I don't see such behavior in 3.0.4.

3.1.0.  It was incorporated as a new feature since it does change
behaviour.

-- 
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