I've got a strange situation here, I have updated to running spamassassin
via spamass-milter with it set to not actually do anything to the message
other than check it and bounce it if it is over a certain score (at the
moment that is set to 50, which should get about 25% of the spam we get
here), I am then running spamassassin on a system-wide per-user basis (I
don't want outbound mail tagged as spam, but I do want the features of the
milter).
Basically what is happening is that the milter is checking the message and
logging one score for the email - and then the per-user spamc process tags
it as either the same score, or different, depending on if it comes through
to an alias or to the real address. I updated the spamd script to do an
alias table lookup and set the correct username to actually do the lookups
correctly (something I would prefer to handle in spamass-milter but to get
things running correctly I added it into spamassassin - I can build a
code-patch if people want it - though this really should not be something
handled by SA as I see the ideals of SA anyhow...).
Basically the question I have is a little more to do with spamc or the perl
interface to SA (I don't know how spamass-milter talks to SA yet - not been
that hassled on it), but if it uses spamc with the -u does that actually
check the @GLOBAL entries in the database as well as the users entries?
Because that seems to be the closest suggestion I have as to why this is
giving differing scores - if it is not using my system defaults for some
score levels but using the SA defaults...
Any ideas on this one would be appreciated.
Regards,
Cassandra
-------------------------------------------------------
This SF.net email is sponsored by: eBay
Get office equipment for less on eBay!
http://adfarm.mediaplex.com/ad/ck/711-11697-6916-5
_______________________________________________
Spamassassin-talk mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
