On 8/18/2016 12:16 PM, John Hardin wrote:
On Thu, 18 Aug 2016, Jerry Malcolm wrote:
I installed the latest SpamAssassin In a Box yesterday (Win Server
2008 r2). I kept all of the defaults. It is up and running. But I'm
getting a huge amount of spam, and I mean 'obvious' spam mentioning
body parts in the subject line.... that are getting low scores
(averaging about 15 uncaught spams per hour per user inbox). It's
still catching some spam. So I assume it just a scoring issue.
I tried running sa-update. It said no updates were available. I'm
fairly certain that SA is better at recognizing spam that it is
currently doing on my system. Is there something else I need to do
in order to get it to begin recognizing obvious spam? Hopefully 15
uncaught spams per hour is not considered the acceptable norm.
Be aware that Bayes scoring doesn't kick in until you've actually
provided some training to allow it to recognize your particular email
traffic.
There are also potential DNS issues that may contribute. In addition
to describing your environment, perhaps you could post the
X-Spam-Status header from a couple of the low-scoring spams.
John,
This is the X-Spam-Status header I got back on an uncaught spam. No,
hits=0.3 required=5.0. The spam was selling an all-in-one charger.
What kind of DNS issues? I lease a server from Peer1 and use their name
servers.
Thanks.
