[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]


To: dnsop@cafax.se
From: "Eric A. Hall" <ehall@ehsco.com>
Date: Wed, 12 Nov 2003 06:59:02 -0600
Sender: owner-dnsop@cafax.se
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.5) Gecko/20031007
Subject: the least-worst default


Of the three classes, the well-known address approach is the only one that
can work in all cases. DHCP and RA both have natural uses in specific
situations but difficulties in other situations, making them poor choices
as the "default".

OTOH, the well-known-address approach *can* work for everybody by default,
at least if a couple of minor changes are made. In particular, it needs to
use a multicast address rather than a unicast address -- servers won't be
able to listen on a unicast address that isn't configured in the OS, but
they should be able to bind and listen to a standardized multicast address
without any problems. And it should probably be restricted to an optype
that chooses a server, instead of being used for all queries -- this is
particularly important given the potential for spoofed answers if ad-hoc
servers are going to be allowed. Those changes can be debated separately,
of course. The important thing here is that this is the only approach that
has the potential for 100% coverage, and thus the only approach that's
suitable as the default.

-- 
Eric A. Hall                                        http://www.ehsco.com/
Internet Core Protocols          http://www.oreilly.com/catalog/coreprot/


#----------------------------------------------------------------------
# To unsubscribe, send a message to <dnsop-request@cafax.se>.

Home | Date list | Subject list