Re: [dhcwg] Comments on draft-ietf-dhc-3315id-for-v4-03

On Aug 11, 2004, at 3:43 PM, David W. Hankins wrote:
> I think we should just define the sane and rational behaviour we
> expect of the server when it encounters the mixed identities, since
> we already appear to have a consensus on what that should be.

David, in the case you are describing, the DHCP client has no stable 
storage.   So there's no benefit to implementing the new client 
identifier type on that client.   The server's behavior towards clients 
that are out of compliance with the draft but do follow RFC2131/2132 is 
clearly stated in the draft.

A host operating system that has been booted by a PXE boot loader that 
does not comply with this draft, along with the PXE boot loader itself, 
together represent a system which can never be in compliance with this 
draft.   The host operating system should use the same identifier that 
the PXE client uses, whatever that is, and not make up its own.

If the host operating does in fact make up its own identifier, then we 
are talking about two different DHCP clients, one of which follows the 
spec, and one of which does not.   The two clients will get two 
different IP addresses.   There is no harm in this, unless you are 
concerned about excessive IP address consumption.   If you are 
concerned about that, there are stopgap solutions to get you over the 
hump until the last of those PXE cards dies a well-deserved death.

While I sympathize with the issue you are raising, before you go any 
further I would suggest that you read the problem statement, and then 
sit down and see if you can write text that resolves the issue you are 
raising in any better way than the current draft does, while at the 
same time not failing to meet the goals of the current draft.   When 
you haven't tried yourself to write the text that addresses a problem 
you're describing in the context of the draft to which you wish to add 
the text, it's easy to imagine that there is some solution that could 
be grafted onto any particular document.   It is by such mistakes that 
documents get delayed, sometimes forever, in their journey through the 
standards process.

_______________________________________________
dhcwg mailing list
dhcwg@ietf.org
https://www1.ietf.org/mailman/listinfo/dhcwg