RE: [dhcwg] dhc WG last call on draft-ietf-dhc-proxyserver-opt-00

Perhaps this has been discussed in the past and discarded, but what
about removing the need for yet-another-list? If the option where
changed not to use sub-options at all but instead carried tuples of:
	2-octets of well-known protocol number
	4-octets of proxy server internet address
	2-octets of proxy server port number
The option length would then be multiples of 8 bytes.

For example:
	Option Header:	TBD-option-code,24-bytes
	Proxy for HTTP:	80,http-proxy-server-1,8080
	Proxy for HTTP:	80,http-proxy-server-2,8080
	Proxy for FTP:	21,ftp-proxy-server-1,21

This avoids the need for future protocols to have an enumeration and the
current "getservicename" calls can be used to find the port numbers.

Otherwise, I recommend moving this draft forward.

- Bernie
-----Original Message-----
From: dhcwg-admin@ietf.org [mailto:dhcwg-admin@ietf.org] On Behalf Of
Ted Lemon
Sent: Wednesday, April 07, 2004 1:57 PM
To: dhcwg@ietf.org
Subject: Re: [dhcwg] dhc WG last call on
draft-ietf-dhc-proxyserver-opt-00

I have a bunch of suggested changes to this draft, which fall into 
three categories:

1. A single change to the protocol itself.
2. Editorial changes - I think there was some extra text in this draft 
explaining proxy servers that isn't needed, and generate questions 
during the IESG review, so I'm suggesting that it be deleted.
3. Copyediting.   There were some minor spelling and grammatical errors.

The change to the protocol is that it currently specifies an 
encapsulation of suboptions, like option 82, but allows for the 
appearance of multiple suboptions, which is different than the behavior 
specified for handling options in RFC3396.   This is not a huge 
problem, but it probably requires additional code in DHCP servers and 
clients that isn't necessary, so I'd suggest changing it so that if you 
want to specify multiple proxy servers for the same protocol, you 
should just list more than one IP address/port tuple in the suboption 
for that protocol.

I've enclosed a diff for all the changes.

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