[NNTP] Proposed changes to draft-ietf-nntpext-streaming-02

[Ken Murchison]

Russ Allbery wrote:

> The proposed changes to all three of the drafts look good to me.
> 
> Ken Murchison <ken at oceana.com> writes:
> 
> 
>>Outstanding issue:
> 
> 
>>    o Do we want to add any discussion of the server issuing a 400 (or
>>      maybe a 403) response to TAKETHIS?
> 
> 
> I think so.  Here's some suggested text:
> 
>     If the server encounters a temporary error that prevents it from
>     processing the article but does not want to reject the article, it
>     MUST reply with a 400 response to the client and close the connection.
> 
> I'm nervous about the effect of 403 on existing implementations, but
> probably wouldn't object strongly if it were added as an additional
> alternative here, like:
> 
>     If the server encounters a temporary error that prevents it from
>     processing the article but does not want to reject the article, it
>     MUST reply with either a 403 or a 400 response to the client.  A
>     400 response (which is followed by the server closing the connection)
>     is recommended if the error is likely to affect all subsequent
>     attempts to transfer articles as well.


I'll hold off on adding text about 403 until we hear from Andrew on 
this.  If we don't discuss 403, I might combine your two versions into 
something like:

"If the server encounters a temporary error that prevents it from
processing the article (such as disk full) but does not want to reject 
the article, and the error is likely to affect all subsequent attempts 
to transfer articles, it MUST reply with a 400 response to the client 
and close the connection."


The point being that if the error only affects the given article 
(however unlikely), closing the connection seems unfriendly.

-- 
Kenneth Murchison     Oceana Matrix Ltd.
Software Engineer     21 Princeton Place
716-662-8973 x26      Orchard Park, NY 14127
--PGP Public Key--    http://www.oceana.com/~ken/ksm.pgp