On Wed, 4 Jun 2008, Obantec Support wrote:
i looked over the above and my server seems to conform but it still scores
low on an example email.
X-Spam-Checker-Version: SpamAssassin 3.2.4 (2008-01-01) on my.mailserver.net
X-Spam-Level: ***
X-Spam-Status: No, score=3.6 required=4.5 tests=ANY_BOUNCE_MESSAGE,AWL,
BAYES_99,BOUNCE_MESSAGE autolearn=no version=3.2.4
VBOUNCE is not intended to mark bounces as spammy by itself, it's intended
to _identify_ them. In your delivery chain post-SA you'd look for
ANY_BOUNCE_MESSAGE in X-Spam-Status and then either deliver to a "bounces
for review" folder, or drop the message.
You could, however, add a meta-rule that adds points for messages hitting
both ANY_BOUNCE_MESSAGE and BAYES_99, if you trust your bayes. I'd say
that's a pretty good indicator of a bounced spam.
Perhaps:
meta BOUNCED_SPAM (ANY_BOUNCE_MESSAGE && BAYES_99)
score BOUNCED_SPAM 4.0
--
John Hardin KA7OHZ http://www.impsec.org/~jhardin/
[EMAIL PROTECTED] FALaholic #11174 pgpk -a [EMAIL PROTECTED]
key: 0xB8732E79 -- 2D8C 34F4 6411 F507 136C AF76 D822 E6E6 B873 2E79
-----------------------------------------------------------------------
A sword is never a killer, it is but a tool in the killer's hands.
-- Lucius Annaeus Seneca (Martial) 4BC-65AD
-----------------------------------------------------------------------
14 days until SWMBO's Birthday
