Hi Hamish,
I have a spam IMAP folder foreach user which he can use to move spam messages to there, and it is used by sa-learn everynight in order build the spam database.
A sample of my logs with a spam detection are: spamd[2712]: connection from localhost [127.0.0.1] at port 56773 Dec 10 09:54:27 xxxxxxx spamd[25604]: info: setuid to jreis succeeded Dec 10 09:54:27 xxxxxxx spamd[25604]: processing message <116638626203.ZNM92475@axn-asia.com>crewmen.freemessage.com> for xxxx:500. Dec 10 09:54:27 xxxxxxx spamd[25604]: identified spam (5.1/5.0) for xxxx:500 in 0.3 seconds, 2428 bytes.
As you can see i do not have as much as information as you have and probably my spamassasin is not as good configure as yours, and i am afraid this is the problem, because i have already marked a lot of LOTERY spam messages in order to sa-learn to learn from it and i continue to receive this kind of messages. Can you point me some direction on how to check my scores and to change its limits?
Thanks
OK you should be able to add this to your conf file: rewrite_header { subject } ***SPAM*** This is OFF by default so that could be your problem. Have a look here for the full details: http://spamassassin.apache.org/full/3.0.x/dist/doc/Mail_SpamAssassin_Conf.ht... Also, to give you a bit more work, but save you some in future... Try out amavisd-new. It is really really cool - It allows you to tie in clamav,spamassasin,a huge number of other scanners and loads of other options in one config file. Here is a nice howto: http://infolinux.de/blog/?q=title/sieve Hope that helps, H