On 5/12/2011 3:22 PM, Manfredi, Albert E wrote:
Mark Smith wrote:
I think it would be reasonable to make DHCP a SHOULD, however
I've thought that one of the reasons SLAAC exists is to provide
simpler and lighter weight address configuration method for resource
constrained end-nodes such as embedded ones. So perhaps it could be
worth mentioning that an example of an exception to the SHOULD would be
those types of end-nodes.
More generally, I'd say SLAAC exists because with IPv6 it's possible to assign
globally unique IP addresses to hosts, without risk of collisions, without
having to have the network specify all of the 128 bits.
But in situations where the address provided to a host must be completely
predictable, SLAAC won't work.
I don't think that resource-constrained need to be a significant criterion?
Rather than trying to enumerate all the reasons that it might be one or
the other, how about language to the effect of:
If building a special-purpose device for use on a network where only
RA/SLAAC is allowed then implementors MAY omit DHCPv6 support. Otherwise
DHCPv6 SHOULD be provided.
Doug
--
Nothin' ever doesn't change, but nothin' changes much.
-- OK Go
Breadth of IT experience, and depth of knowledge in the DNS.
Yours for the right price. :) http://SupersetSolutions.com/
--------------------------------------------------------------------
IETF IPv6 working group mailing list
ipv6@ietf.org
Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------
