On 2008-12-10 at 17:49 +0100, Marcin Krol wrote:
> I also would like to implement selective greylisting, for mails that are
> "suspicious" (like when SA score is above SAtempreject in sa-exim), and
> this would be much easier to do with external script than implementing
> it all in Exim.
Indeed; I use such a program, the greylisting program from Debian,
written in Python and hacked around to run on my non-Linux system.
It uses a unix-domain socket for communications, I communicate with it
via the ${readsocket...} expansion.
Myself, the idea of letting each inbound mail, prior to filtering out
the spam, fork a process worries me. Exim goes to some lengths to
provide load limitation controls and forking new processes during the
earliest stages seems fraught with problems.
However, if I did want to do that, I'd use ${run...} as an expansion.
Carefully.
defer message = $sender_host_address is not yet authorised to \
deliver mail from <$sender_address> to \
<$local_part@$domain>. Please try later.
log_message = greylisted
[... various checks, including whitelisting on file and dnswl.org ...]
domains = +local_domains
verify = recipient
condition = ${readsocket{/var/run/greylistd/socket}\
{--grey \
$sender_host_address \
$sender_address \
$local_part@$domain}\
{5s}{}{false}}
-Phil
