On 19 Jul 1999, Vladimir Litovka wrote:
> > However, Exim's basic philosophy is that if a host has been down for a
> > very long time, there is no point keeping messages hanging around.
>
> What is very long time?
Exim doesn't define this itself. It is the maximum retry time for the
host in question. It can of course be quite short, but when I
implemented it I was thinking of conventional mail, and times of 4 or 5
days. If a host has been down for 4 days, you might as well tell a
message's sender right away instead of carrying on trying.
> In my case, for example - it is mail for cellular
> operator's clients, and this mail becoming unnecessary for recipients
> very quickly - in few hours, not more. So in some cases mail growing old
> itself, and this process doesn't depend on server's down time.
The short answer (of course) is that Exim wasn't defined with this kind
of usage in mind. Perhaps what you should do is build a special-purpose
router to handle this. Create a queryprogram router which inspects the
message (you may have to pass the message id and have it inspect the
date of creation of the -D file on the spool, though maybe the contents
of $h_date: might do). If the message is young, just return FAIL, and
the message gets passed to the normal routers for delivery, or deferring
as now. If the message is too old, return OK and pass it to a /dev/null
transport.
I haven't tried any of this. It is just an idea off the top of my head.
--
Philip Hazel University of Cambridge Computing Service,
ph10@??? Cambridge, England. Phone: +44 1223 334714.
