431 for TAKETHIS

Pavel Gulchouck gul at lucky.net
Thu Sep 16 15:28:19 UTC 1999


On Thu, Sep 16, 1999 at 04:35:22PM +0200, Fabien Tassin writes:
> According to Pavel Gulchouck:
> > 
> > Is response 431 correct for TAKETHIS command?
> 
> I don't think so because TAKETHIS is inconditionnal (the article is just sent).

But answer "436" is correct for IHAVE when the article is already
transferred.
Article is sent only after receiving corresponding answer from remote.
It's poor if receiver must drop NNTP-session if it cannot get the
article now but want to do it later, and it didn't know about this
while responds to "CHECK" (for example, it can depends of "Newsgroups:"
header, of article size or of network or sendmail state).

Comment in innfeed sources said the same:
> > case 431:             /* try again later (also for TAKETHIS) */
and fix can be makes by small source changes (about 5 lines).

> > Btw is anywhere a complete specification for NNTP streaming extension?
> 
> yes. Look at the latest IETF draft called draft-ietf-nntpext-base-08.txt

I don't found anything about stream mode there. :(

# grep -i takethis draft-ietf-nntpext-base-08.txt
#

> -- 
> Fabien Tassin -+- fta at oleane.net

-- 
                                Lucky carrier,
                                                  Pavel.



More information about the inn-bugs mailing list