Firstly thanks to everybody who so quickly replied to my problem. It seems when it rains it pours and I have been tied up with several other probs at the moment and haven't been able to respond. (what a great mailing list though)
This is a example: snip... [EMAIL PROTECTED]>, Hits: -3.629 May 13 13:55:56 mailserver amavis[10743]: (10743-05) TIMING [total 959 ms] - SMTP EHLO: 2 (0%), SMTP pre-MAIL: 0 (0%), SMTP pre-DATA-flush: 3 (0%), SMTP DATA: 30 (3%), body hash: 1 (0%), mime_decode: 31 (3%), get-file-type: 20 (2%), get-file-type: 16 (2%), decompose_part: 2 (0%), decompose_part: 0 (0%), parts: 0 (0%), AV-scan-1: 540 (56%), SA msg read: 3 (0%), SA parse: 2 (0%), SA check: 196 (20%), fwd-connect: 9 (1%), fwd-mail-from: 2 (0%), fwd-rcpt-to: 2 (0%), write-header: 6 (1%), fwd-data: 1 (0%), fwd-data-end: 85 (9%), fwd-rundown: 2 (0%), unlink-2-files: 5 (1%), rundown: 0 (0%) I am using SpamAssassin-2.63 with amavis new. Jonas Eckerman kindly wrote: >Are you using bayes? Did the bayes database "suddenly" reach the required 200 hams and 200 spams? >If yes: Negative scores are perfectly normal when using bayes. That's the whole point of using it. Yes I'm using bayes but am not sure what the command would be to see how many are in my bayes database ? It could be that because I have had a large amount of users using this system for email since Saturday onwards. I wrote that I am assuming negative scores are a bad thing. Jonas Eckerman replied: >Why? >Identifying ham (giving them low scores) is just as important as identifying spam Very happy to hear this if this is the case. Sounds like than I don't have a problem and spamassassin is running sweetly. Anybody have any views to share over this last point ? I didn't know spamassassin was 'smart' enough to do this. Thanks again to everyone !!!
