Re: capitalization was Re: AXFR "clarify"

[Mark Andrews <Mark_Andrews@isc.org>]

> At 12:34 +1100 1/19/08, Mark Andrews wrote:
> 
> >	I said it was a general issue.  People however worry about
> >	it more with AXFR than with other queries as AXFR provides
> >	a much larger set of compression targets.  It gets realised
> >	rather than being potential.
> 
> In that case, it should be mentioned as something to think about when 
> writing code but does not place any requirements on the AXFR 
> mechanism.

	We are in this mess because people didn't apply the general
	requirements of RFC 1034 to AXFR.

	* message id's matching is a general requirement.

	* setting flag bits correctly is a general requirement.

	* setting rcode correctly is a general requirement.

	* that the answer is in the answer section and not the additional
	  or authority sections is a general requirement.

	* not mixing zone content is a general requirement.

	* preserving case is a general requirement.

	* multiple records that compare equal are treated as a single
	  record.

	If we don't want to re-iterate general requirement then we
	are done.

	The only thing that is special about AXFR is that it is
	restricted to a reliable streaming protocol (e.g. TCP) and
	that because there wasn't a general mechanism for splitting
	answers over multiple DNS messages a query specific framing
	method was required to identify when the AXFR was complete
	(SOA at start and end).  Strictly speaking only the final
	SOA is needed.

	Mark
-- 
Mark Andrews, ISC
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: +61 2 9871 4742                 INTERNET: Mark_Andrews@isc.org

--
to unsubscribe send a message to namedroppers-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://ops.ietf.org/lists/namedroppers/>