How long should a client wait for a DHCPOFFER?

David W. Hankins David_Hankins at isc.org
Fri Oct 13 19:09:36 UTC 2006


On Fri, Oct 13, 2006 at 02:44:35PM -0400, Aaron Bennett wrote:
> http://www.faqs.org/rfcs/rfc2131.html
> 
> I think the relevant section is:
> 
> >      algorithm in section 4.1.  The client should choose to retransmit

Unfortunately, 4.1 isn't very specific.  It only 'MUST's that the
client use an exponential backoff, and provides an example for "a
reasonable 10base ethernet approach".

So implementors are free to choose whatever methods they wish to.


Empirically, generally all clients I'm aware of use a 3 second
timeout initially, and exponentially backoff from there, with
random jitter up to 1/2 the backoff step.  This is loosely modeled
after rfc2131, as the best you can come away from section 4.1 is
to think of the minimum initial timeout as 3, and the intention
of the rest of the text is that DHCP via long-distance radio to Mars
or similar should probably use substantially longer timeouts.

I don't think this operational choice is documented anywhere.

-- 
ISC Training!  October 16-20, 2006, in the San Francisco Bay Area,
covering topics from DNS to DDNS & DHCP.  Email training at isc.org.
-- 
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


More information about the dhcp-users mailing list