Magnus Holmgren wrote:
On Wednesday 14 February 2007 14:55, maillist wrote:
Content analysis details: (8.6 points, 7.0 required)
pts rule name description
---- ----------------------
--------------------------------------------------
2.4 SPF_HELO_SOFTFAIL SPF: HELO does not match SPF record (softfail)
[SPF failed: Please see
http://www.openspf.org/why.html?sender=janis.com&ip=212.11.121.229&receiver
=mail.emailacs.com] -1.8 ALL_TRUSTED Passed through trusted hosts
only via SMTP 8.0 BAYES_99 BODY: Bayesian spam probability is
99 to 100% [score: 0.9970]
Is there any reason that such a message with the above score would make
it to an in-box?
That depends entirely on whatever moves the spam to the spamdrop in-box. What
MDA do you use and what criteria is it configured to make its decisions upon?
Sorry I should have giving my config. Here it is:
OS - slackware 11.0
SpamAssassin Server version 3.1.7
running on Perl 5.8.8
with SSL support (IO::Socket::SSL 1.02)
sendmail Version 8.13.8 - MBOX format
mimedefang version 2.58
I have mimedefang set to move all spam to a spamdrop in-box.
All of my users are getting these messages a few times
a day. Other than that, all other spam is correctly moved to a spamdrop
in-box. I sent a question in the other day about this, and never heard
anything back from anyone. I'm still puzzled by this. I don't have any
sort of whitelist setup.
Does any of the correctly moved spam have a lower score than this?
Yes. Just glancing at the spamdrop in-box, I see 1 message that scored
8 and that was only due to BAYES_99
Content analysis details: (8.0 points, 7.0 required)
pts rule name description
---- ----------------------
--------------------------------------------------
0.0 HTML_MESSAGE BODY: HTML included in message
8.0 BAYES_99 BODY: Bayesian spam probability is 99 to 100%
[score: 1.0000]
....It's baffling.
