On 8/12/2014 10:05 AM, Matus UHLAR - fantomas wrote:
On 8/12/2014 6:31 AM, matth wrote:
Please have a look at the email below: in the content analysis report (in
the body) the spam score appears as 5.1 points. The email is correctly
identified as spam, the subject line changed to include "*****SPAM*****".
However, in the email headers the score appears as 0.001, message is not
marked as spam, and even considered a good example of ham! (autolearn=ham).
I really can't understand what's going on... would appreciate any
assistance.
On 12.08.14 09:13, Bowie Bailey wrote:
Double processing. The first pass through SA marked it as spam
actually, not SA, but amavis. Yes, amavis uses SA engine but it's not SA.
I didn't look at it that closely. The point stands. There is a problem
with his mail processing path that is causing the message to be scanned
twice.
Perhaps Amavis is scanning the message on receipt and then SA is being
called directly during delivery.
--
Bowie
