I've been adding extra headers to incoming emails detailing everything
from source port range to the passive OS fingerprint in hopes that the
bayes database or my rules will pick them up and assign additional (or
reduced!) spam scores for them.
I haven't had the chance to truly experiment and see if it is helping or
not, but something just occurred to me (at 3am as I was going to bed).
Do the headers I add with "add_header" actually get placed into the
message which is passed to SpamAssassin and thus are available to learn
and score, or are they only added during delivery in the transport?
If not, how on earth would I pass such information to SA so it can be of
use? I am currently calling SA in the acl_smtp_data ACL.
http://www.exim.org/exim-html-current/doc/html/spec_html/ch40.html#SECTaddheadacl
From what I read in the above section, I can only guess that if I add
headers in the MAIL, RCPT and pre-DATA ACL, they _may_ be available in
the message sent to SA. Has anyone played with this at all?
Ted.
--
The Exim Manual
http://www.exim.org/docs.html
http://www.exim.org/exim-html-current/doc/html/spec_html/index.html
