Re: [dnsext] Re: I-D ACTION:draft-vandergaast-edns-client-ip-00.txt

[Joe Abley <[e-mail]>]

On 2010-01-29, at 14:45, Paul Hoffman wrote:

> Section 4.1 says "An edns-client-ip option MUST never contain a non-public address." To me, that means two things:

Ah, I missed that text, thanks.

> The biggest stumbling block here is the assumption that developers will know the correct list of non-public addresses. Given lots of deployed counter-examples, that seems like a bad assumption. Specific pointers to specific IANA registries are needed in this draft.

Yes, it does seem to imply the hard-coding of special values into software.

An alternative would be to add the query source as viewed by every option-understanding nameserver that receives the query and performs a recursive lookup, forget the "non-public" text in 4.1 and let any nameserver that whats to process the option make whatever interpretation they want. That isolates the need to concern yourself with unique vs. non-unique addresses to the point of analysis, not data collection.

I think the general idea of being able to better understand your user community would have some operational/diagnostic benefits for any authority-only nameserver. No doubt there are also privacy implications.


Joe