Re: [dhcwg] Leasequery: should it be standardized?

> My recollection, though, was that back in the beginning of all this, 
> DHCPLEASEQUERY was about providing a way to refresh the ARP cache of a 
> concentrator without having to broadcast an ARP request to every one of 
> the concentrator's ports.   Is that no longer the case?

Apparently so, and that is what has prompted my asking questions. The
scope of the problem seems to have expanded quite a bit beyond:

  "providing a way to refresh the ARP cache of a concentrator without
  having to broadcast an ARP request to every one of the
  concentrator's ports",

to something perhaps closer to providing "access control in router
type devices", including cases where DHCP isn't even being used.

I don't have a real problem with leasequery as it was originally
described. But if the scope of the problem being solved expands well
beyond the original problem, it's no longer clear that:

 - the problem statement is adequately well understood and accepted by
   the broader community (or maybe even the DHC community).  Maybe
   other folk not familiar with DHC need a solution to the broader
   problem and will see this solution as being deficient for their
   needs.

 - it's no longer clear whether DHC has the right expertise to get the
   problem defined properly or to work out a good solution.

 - it's no longer obvious that DHC is the best solution to the expanded
   problem.

I am not saying that leasequery as a concept is flawed, but I am
asking questions about the current version of the leasequery draft and
whether it has the right level of functionality and features in it and
is the right thing to standardize on.

Thomas

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