ietf
[Top] [All Lists]

RE: Last Call: 'Linklocal Multicast Name Resolution (LLMNR)' to Proposed Standard

2005-08-31 15:08:09
I'm a by-stander on this discussion, maybe off-base or out of it -- but something other than the undesirable traffic struck me.

Isn't it also true that I might *deliberately break* all sorts of things by introducing 'blocking' names into DNS responses, so that an LLMNR request is never issued. So an ISP could 'grab' traffic that the users thought was local, by replying to a DNS request in a proxy (or converting a negative reply into an answer).

Also, ISPs might be tempted to start turning around DNS requests in their proxies for names that they *think* should be answered by LLMNR, returning resolution failure, so as not to send too much traffic outbound. This pre-empts the real DNS from ever actually replying.

The whole idea that 'real DNS' can arbitrarily pre-empt local name resolution seems, well, wrong, and needs serious study for security implications for the services using those names, no?

--
David Singer
Apple Computer/QuickTime

_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf

<Prev in Thread] Current Thread [Next in Thread>