At 8:11 PM +0300 9/23/09, Jari Fredriksson wrote:
> On Wed, Sep 23, 2009 at 07:07:53PM +0300, Jari
Fredriksson wrote:
Jari Fredriksson wrote:
Then I decided SaneSecurity is not worth it, as
SpamAssassin catches those too, and has less false
positives.
SaneSecurity triggers way too often when some dumb user
pastes a spam into his mail, or some robot sends a
bounce with an attachment. I do not want to report
those cases to SpamCop, Razor, DCC.. Making me writing
tons of tests in my scripts. Too risky.
If someone pasts a spam into their mail it is not a
false positive. It is a post that is indistinguishable
from spam. There are consequences for that.
Debatable. Anyway, I do not want to punish from that
kind of a mistake. I'm not an email nazi, while I indeed
am a spam fighter.
Ehm, were you scoring SaneSecurity hits like one is
supposed to, or just plain rejecting with them? Sounds
like the latter.
I don't run ClamAV via SpamAssassin. I have it called by
amavisd-new, which does what it does: quarantine.
Sure hope your not using heuristics, phishing and/or safebrowsing
options in ClamAV if you feel that way.
Tom
_______________________________________________
Help us build a comprehensive ClamAV guide: visit http://wiki.clamav.net
http://www.clamav.net/support/ml
