ietf-nntp New wording on article numbers - draft 3

[Jack De Winter]

>>Any ideas on something similar?  And how do we get around the backfill
>>problem with parallel feeds to a single group?  We have to handle it
>>in some way that everyone likes. (My personal feeling is to restrict
>>the number of threads that can feed a newsgroup in this situation.)
>
>Master/Slave replication, especially on large scales, brings up a
>whole can of specialised worms.
>
>My own opinion is that such replication is sufficiently specialised
>to warrant its own protocol, rather than using stripped-down nntp
>servers, Xref: header processing, variations on XREPLIC, or any one
>of a multitude of other local pieces of code.
>
>Would it not make sense to set up some kind of investigative group,
>outside of the actual NNTP WG (though I would imagine that there would
>be a great deal of overlap in the membership), to deal with those
>issues particular to master/slave replication?

It might, but consider the following: if there is something else that
is accessing the news store, should it not also follow the same rules
as we are dictating here?  I mean, would it get into the <X> protocol
that it is okay to backfill the news store, even though the NNTP
protocol, which is different forbids it?

regards,
Jack
-------------------------------------------------
Jack De Winter - Wildbear Consulting, Inc.
(519) 576-3873		http://www.wildbear.on.ca/

Author of SLMail(95/NT) (http://www.seattlelab.com/) and other great products.