Thoughts on new overview

Russ Allbery rra at stanford.edu
Fri Sep 6 03:03:28 UTC 2002


Jeffrey M Vinocur <jeff at litech.org> writes:
> On Thu, 5 Sep 2002, Russ Allbery wrote:

>> I think that what tdx-util does is more correct per the proposed RFC
>> 977 replacement than what innd is doing, although we still haven't
>> hashed out what the byte count is actually supposed to be.

> Hmm.  I'm not sure that the overview data for an article should depend
> on what storage method is being used.  Which is what happens here, if
> rebuilding has occurred.  So regardless of which we use, we probably
> should be consistent, no?

Yup.  So I think tdx-util probably currently does the wrong thing for
non-wireformat articles.  (Which I don't normally think about much, since
I run all my servers with wire format storage.)

The current theory of ietf-nntp is that the byte count should reflect the
size of the article on the wire, since that's the most useful and most
predictable number from the perspective of the client.

It's unclear whether that should include the terminating .\r\n, though.

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

    Please send questions to the list rather than mailing me directly.
     <http://www.eyrie.org/~eagle/faqs/questions.html> explains why.


More information about the inn-workers mailing list