Claudia Burman wrote:
Gary V wrote:
I use spamassassin (last perl version, updated it last week) on a mail
server, called from amavisd-new. I've set the $sa_kill_level_deflt to
5.00, if I lower this I get too many false positives.
I haven't touched any of the rules. I regularly train the bayesian
filter
with false negative messages. I'm using local tests only.
Claudia Burman
Why are you using local tests only?
Gary V
_________________________________________________________________
Don’t just search. Find. Check out the new MSN Search!
http://search.msn.click-url.com/go/onm00200636ave/direct/01/
Well, I have enabled network tests and the server load got too high,
so I disabled razor and pyzor. Now server load is acceptable and spam
caught is about 95%!
Thanks to all.
Claudia
For what it's worth on my system far less than 1% of spam gets through.