Re: [dhcwg] dhc wg last call on "DHCP Relay AgentAssignmentNotification Option"

The protocol is defining a real solution for which DHCPv4 "snooping" or
"gleaning" was developed as a workaround.

There are many use cases for getting state into the network element on which
a relay agent runs based on DHCP activity.  This state could be transmitted
through some other protocol; e.g., TELNET/CLI to the network element, SNMP.
It seems the most direct method to deliver the state information to the
network element is to piggy-back that information on the DHCP message
exchange.

We missed the opportunity to formalize this information delivery in DHCPv4.
This protocol for DHCPv6 will formalize the specification, add reliability
and more closely follow the DHCPv6 model of client-server message
encapsulation (in constrast with the DHCPv4 model of message
forwarding/relaying).

- Ralph

On 6/14/06 10:30 PM, "Ted Lemon" <Ted.Lemon@nominum.com> wrote:

> Thomas Narten wrote:
>> 
>> I do think we need to make our protocols fairly rock solid. Given how
>> this is core infrastructure, it just has to work. All the time.
>>   
> To be clear, I was questioning the need for the protocol for which
> Bernie's fix applies, not the need for the fix.   If we adopt the
> protocol, we have to have Bernie's fix.   But I'm not convinced we
> should adopt the protocol.

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