Re: repeating records

[gson@nominum.com (Andreas Gustafsson)]

D. J. Bernstein writes:
> Andreas Gustafsson writes:
> > Each of them SHOULD be transmitted only once
> 
> That blatantly violates RFC 2119, section 6.
> 
> > > Explaining that servers can repeat records, and requiring clients to
> > > handle repetitions, would be a clarification of the protocol.
> > That's what the draft says
> 
> And I have no objection to the draft saying that. What I object to in
> this area is the draft saying more---this ludicrous BIND-8-violating
> ``clarification'' that records ``SHOULD be transmitted only once.''
> According to RFC 2119, this means that
> 
>    there may exist valid reasons in particular circumstances to ignore a
>    particular item, but the full implications must be understood and
>    carefully weighed before choosing a different course
> 
> which nobody claims is true in this case.

I would certainly want implementers to understand and carefully weigh
the implications of multiple transmission, such as waste of bandwidth
and receiver CPU time.  Are you suggesting they should not?

RFC 2119 section 6 says imperatives (e.g., MUST or SHOULD) "MUST only
be used where it is actually required for interoperation or to to
limit behavior which has potential for causing harm (e.g., limiting
retransmisssions)".  This would be the latter case.  Incidentally, I
found it amusing that the author of RFC2119 specifically chose
limiting retransmissions as the example; although he probably had a
different kind of retransmission in mind than that of transmitting
records multiple times in an AXFR, they both cause similar harm
in the form of wasted bandwidth.
-- 
Andreas Gustafsson, gson@nominum.com

--
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/>