On Fri, 21 Jan 2005, Tore Anderson wrote:
> Right before writing the message out to the spool, I'd guess? Right
> before the DATA ACL, in other words. Or do you need access to spool
> directory before that?
You do if logs are being written there (for example). Or are we talking
about a sub-spool for just the "input" directory? I think perhaps we
are. So this isn't about spool_directory at all, its about providing a
parameter for "input". In some ways, that might be easier.
> However, the possible problem I see is only with immediate
> deliveries, because as you point out, the immediate delivery process
> doesn't neccesarily know where to find the message the daemon told it
> about with -Mc.
Or the human told it about, or the exim monitor told it about...
> I'd also need a dedicated queue runner process for this spool
> directory, of course. Similarily, this would just use the -SD
> parameter, for instance "exim -q10s -SD/var/spool/exim/vodafone".
> You'd also need the -SD paramter for other commands such as -bp,
> -Mvl, -R, et cetera. Unless you're just using the default set in the
> main configuration, making the whole thing backwards compatible.
There would be a demand for the daemon to start queue runners for the
different queues.
> You still think it's impossible? Why? (Surely I must've thought
> of everything!) :-)
Now that I have reduced it to an alternate input directory rather than a
complete alternate spool, I think it is impossible on a slightly lesser
scale. :-) Remember, there are other things such as retry data and stuff
in the spool directory. And message logs.
This would be a Big Project, if only because the entire source would
have to be combed for references to spool_directory, and modified in a
number of different ways.
I don't see myself doing any Big Projects this year -- major work will
be to get the doc sources and test suite into a standard format that
others can use. There are also at least two other Big Projects that I
can think of which have been waiting a while: re-organizing header
management in ACLs and abstracting SMTP delivery so that callout can use
the same code. There are probably more that I've forgotten...
--
Philip Hazel University of Cambridge Computing Service,
ph10@??? Cambridge, England. Phone: +44 1223 334714.
Get the Exim 4 book: http://www.uit.co.uk/exim-book
