[ietf-nntp] OT: Handling 400 response between NNTP peering servers

Russ Allbery rra at stanford.edu
Mon May 17 03:27:10 PDT 2004


Nick Brandon <maillist at css-uk.net> writes:

> Is there a consensus on how client-server nntp peers handle the 400
> response. From "draft-ietf-nntpext-base-22", I understand that the
> server can issue this response if it needs to terminate the TCP
> connection.

> Should the client peer wait a random time (say between 1 - 60 seconds)
> and try to reconnect as demonstrated below?

>   [C] IHAVE   <message.id at example.com>
>   [S] 400 Temporarily unavailable
>   (Server closes connection)
>   (Client waits 10 seconds, resolves the server address and tries to
> reconnect)
>   ...

Yup, generally.

> Should there be a "good practice" example included within the NNTP
> protocol?

I'm not sure that this issue is really specific to the NNTP protocol.
It's similar to what one would do with any protocol after an abnormal
termination of the connection by the server.

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



More information about the ietf-nntp mailing list