Everything autolearn=not spam

[John Crossan]

We had a problem with the bays being poisoned.
What we did was give low values to BAYES_00 = -.04 and the like.
By setting bays scores to not subtract much if it thinks it is a low
probability, It helped us.

I hope that you find this helpful.

By the way, Be careful.
The Sober virus is very active, we have been getting a lot of Sober
viruses/worms that we have been blocking and placing in  quarantine the last
couple of days.

John Crossan
Systems Administrator
Valley Presbyterian Hospital

-----Original Message-----
From: MailScanner mailing list [mailto:MAILSCANNER at JISCMAIL.AC.UK]On
Behalf Of Quintin Giesbrecht
Sent: Wednesday, May 04, 2005 9:44 AM
To: MAILSCANNER at JISCMAIL.AC.UK
Subject: Everything autolearn=not spam


My mailscanner/spamassassin box has developed a serious problem.  It is
autolearning everything as not spam.

I have tried deleting bayes and starting over, restored bayes data, and
MS config from last known good backup, upgraded to newest version of MS
(I already have 3.03 of SA).

This is running on Fedora Core 3.

Any help is greatly appreciated!

Thanks,


Quintin Giesbrecht
IT Professional,
Hanover School Division

------------------------ MailScanner list ------------------------
To unsubscribe, email jiscmail at jiscmail.ac.uk with the words:
'leave mailscanner' in the body of the email.
Before posting, read the Wiki (http://wiki.mailscanner.info/) and
the archives (http://www.jiscmail.ac.uk/lists/mailscanner.html).

Support MailScanner development - buy the book off the website!

------------------------ MailScanner list ------------------------
To unsubscribe, email jiscmail at jiscmail.ac.uk with the words:
'leave mailscanner' in the body of the email.
Before posting, read the Wiki (http://wiki.mailscanner.info/) and
the archives (http://www.jiscmail.ac.uk/lists/mailscanner.html).

Support MailScanner development - buy the book off the website!