On Mon, Sep 30, 2002 at 10:29:52AM -0400, Chris Leahy wrote: > 1. A LOT more spam is getting through undetected.
2.4 has a heavy no-false positive bias. The scores in 2.42 should help out a bit. :) (I've seen a few more false negatives with 2.4, but very few...) > 3. Identical messages yield different scores on subsequent sendings. > Sometimes radically different. > ie: a test spam message scores a 5.3, gets labeled as spam and > blocked. An identical test message > one minute later scores 2.1 and gets passed through. Messages should always get the same local-SA score if they're the same message. network-SA may have different scores since RBLs, Razor, DCC, etc, are all outside of SA's control. If you are receiving duplicate messages that score differently for local-only tests, I'd like to see those messages. -- Randomly Generated Tagline: "If you're running the latest version of IIS, then you're not vulnerable to this [security hole], but you're vulnerable to something new." - Phil Cox
msg08254/pgp00000.pgp
Description: PGP signature
