ietf-nntp Requirement for timeout close unclear

Russ Allbery rra at stanford.edu
Sun Jul 8 02:40:44 PDT 2001


greg andruk <gja at meowing.net> writes:
> On Fri, Jul 06, 2001 at 05:07:31PM -0700, Russ Allbery wrote:

>> So the client may have to take that into account if it can't reuse the
>> message ID.  I don't see that as much of a problem; basically, it just
>> says "don't retry right away in the rare case where you didn't get an
>> affirmative response from the server unless you can generate your own
>> message IDs."

> The only real hang-up is that some newsreader writers are reluctant to
> build IDs on their own.  This can be helped by turning the MsgID-in-340
> response into a formal extension.  Posting time may be too late for some
> clients to make good use of that information, so an alternative might be
> to use the GETSET extension.

Agreed.  Until then, I stand by my proposed wording, and would argue that
that's why it says SHOULD and not MUST.

-- 
Russ Allbery (rra at stanford.edu)             <http://www.eyrie.org/~eagle/>



More information about the ietf-nntp mailing list