Is sa-learn not learning?

Hello, I am trying to get sa-learn to work, but I can't tell if it is actually doing anything because although I read that it knows which messages it has already scanned, everytime I run it, all the messages are scanned. What am I doing wrong? Here is what happens when I run sa-learn 3 times i

Confused about which bayes db gets used with spamc?

Hello, I run spamc from my procmail on incoming messages. Does this mean that all messages are using root bayes_db? If so why do the clients have stuff updated in their db in their home directories? I am trying to figure this out so I can do sa-learn correctly. Thanks, CP -- View this message