Re: well-known addresses / was DNS discovery

["Eric A. Hall" <ehall@ehsco.com>]

Mark.Andrews@isc.org wrote:

> 	It's not a requirement if the ONLY thing you want clients
> 	to know about the DNS is the addresses of nameservers.  In
> 	the real world there are lots of organisations that want
> 	to push other configuration details out.  The proposals
> 	you are competing with can support this.

Well... It's 'possible' to pass a search list to the resolvers as part of
a startup negotiation, it's just not necessary, and it may even be
determined as undesirable after some more thinking.

In particular, if the logic focuses on choosing a candidate server versus
using any available server, then the server part of the algorithm can
require "return a list of SOAs for which you are authoritative", and the
client algorithm can allow the resolver to use the owner names of the SOAs
as the search list. Extending this logic a bit, it's feasible that any
particular implementation could provide a configuration ~directive that
filtered the domains which the server should return, thereby providing a
way for managers to control the search list array in use by the clients.

Negotiation versus queries isn't the focus of the current WKA proposal, of
course, but as I already said, I think that's the second necessary change
(switching to multicast is the first; anycast is a non-starter).

-- 
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>.