On Wed, 6 Dec 2006, Chris Lightfoot wrote:
> I was considering the case of the system which is
> attempting final delivery detecting an over-quota
> condition and reporting it to a connected MTA as a
> permanent failure, rather than that of the connected MTA
> having seen 4xx errors continuously for a long time and
> reporting a 5xx quoting the text of the error. The latter
> is obviously behaving sensibly.
I'm not sure why I got into this discussion, particularly as I've
actually been talking nonsense in describing what we do. We send neither
4xx nor 5xx for over quota because we don't check at SMTP time.
However, looking at the above, how do you distinguish it from the final
delivery system reporting with 4xx for some time, and then when it's
decided that the user isn't going to fix things, changing to 5xx (for
that user)? I think my general feeling is that changing anything
involves far too much second-guessing.
--
Philip Hazel University of Cambridge Computing Service
Get the Exim 4 book: http://www.uit.co.uk/exim-book
