when is Bayes scoring used?

Julian Field mailscanner at ecs.soton.ac.uk
Wed May 7 13:50:17 IST 2003

At 13:32 07/05/2003, you wrote:
>OK - this is really getting a little confusing...
>I checked my "learn.spam.log" this morning and a found the following entries:
>Wed May 7 00:01:01 EDT 2003
>Learned from 4 messages.
>Learned from 1 messages.
>Wed May 7 01:01:01 EDT 2003
>The problem is that when I run sa-learn -D --rebuild I still get the
>message that says:
>Only 87 spam(s) in Bayes_db < 200
>(it should be AT LEAST the 87 form yesterday plus the ones listed above -
>Can anyone tell me how to fix this? The Bayes files on /root/.spamassassin
>are all being updated multiple times per day so I know it is working,
>unless of course the sa-learn command is reading Bayes info from another
>directory that really DOES only have 87 spam(s).
>Is there a way to run sa-learn and have it tell you the path that it is
>reading the Bayes info from?

Have you tried
         sa-learn -D
? I just ran "sa-learn" on its own and it prints the usage for you.

>Thank for any assistance.
>At 02:38 PM 5/6/2003 -0400, you wrote:
>>something else to add...
>>According the script that Julian provided to run sa-learn through cron,
>>my log is called "learn.spam.log"
>>When I checked that file - I added up all of the "learned form XX
>>messages" and the total number was 447.
>>Is the "learned from" referring to spam and ham? Is it possible that I
>>have 87 spam and the rest of them a ham? I thought I was pretty sure that
>>more spam was getting processed than ham - but I could be wrong.
>>Can anyone shed a little light?

Julian Field
MailScanner thanks transtec Computers for their support

More information about the MailScanner mailing list