I have a question regarding the spam declaration in the data acl.
I have implemented per user scanning where appropriate in our
configuration and if conditions are met will be running the following
warn condition = ${if >{$recipients_count}{1}{1}{0}}
set acl_m1 = 1
---
warn message = X-Spam-Score: $spam_bar ($spam_score)\n\
X-Spam-Score-Int: $spam_score_int\n\
X-Spam-Report: $spam_report
condition = ${if and{{<{$acl_m1}{1}}{<{$message_size}{80k}}}}
set acl_m9 = 1
spam = ${lc:${extract{1}{@}{$recipients}{$value}}}:true
set acl_m9 = 0
We are moving live to exchange2007 as our staff host soon and I have
been able to set up a global filtering rule, such that.
If "custom email header" == 1 then filter to default junkmail folder
I can set this custom header up, eg QUBSpam at our MTAs(exim) using a
mysql statement to extract the users spam preferences from mysql,
however from reading the exim spec the following intrests me
"The spam condition returns true if the threshold specified in the
user’s SpamAssassin profile has been matched or exceeded. "
Does this mean that I could instead of doing another mysql lookup simply
store the result of the "spam = username" declaration and use this as
the conditional as to whether the users preference has been reached or
surpassed!?
R
--
Regards
Ronan McGlue
===================
Analyst / Programmer
Queens University Belfast
