ietf-nntp Multiline Response failure?

[Clive D.W. Feather]

-----BEGIN PGP SIGNED MESSAGE-----

In message <000501c018d7$974fd280$0301a8c0 at tims>, Tim Roberts
<troberts at bigfoot.com> writes
>Another point that I think is not addressed is when a multiline response
>fails in the middle of the response.

>Should it immediately send a ".CRLF" to end the body and then an appropriate
>error response (503)?

A client that doesn't explicitly flush input will read the 503 as
applying to the *next* command.

We have no way to say "this multiline response is wrong", so closing the
connection *without* sending a dot (and this must be clear in the
wording) is the only option.

I would have no objection to an extension that allowed response codes to
come *after* the multiline response or some such mechanism.

- -- 
Clive D.W. Feather    | Internet Expert      | Work: <clive at demon.net>
Tel: +44 20 8371 1138 | Demon Internet       | Home: <clive at davros.org>
Fax: +44 20 8371 1037 | Thus plc             | Web:  <http://www.davros.org>
Written on my laptop; please observe the Reply-To address

-----BEGIN PGP SIGNATURE-----
Version: PGPsdk version 1.7.1

iQEVAwUBObiXhCNAHP3TFZrhAQHEjwf/cYFfr2DcqUnEPw9YcZMDKm2iFehuIzcA
6Uk4077iLHVv0K9Yq7tLkugExRpt2FPbppJaGA472pbOP3wj8sE9O3gVCivBIcp0
sV6yS9KMpFUzyagtzdhOYs/7PKxIiwuXUT70+Y8QsRepySgjXPT1jiJROqj/plk9
pseIan563Vlier31fefsQVXoyxfymYiXxTY0IHTAiV+3K7CiTTzS/fyCcoEtF4NL
bVf+46cEJ9msAZL3w9peh5c8FZyu5sMeEOiu5LZ9h2XpBRJlZbqhy/EKzZU36u7k
b4hzXJcBtQabbh8uzgBb7YNChVX49qR+gjt8x/pZj6qf9XX7EVfj/A==
=l7IS
-----END PGP SIGNATURE-----