Setting negative pontuation can cause false positives?

2010-05-10 Thread Lauro Costa G. Borges
SpamAssassin version 3.2.5 running on Perl version 5.10.0 Ubuntu 8.10 I configured some addresses to be -9.0 or even -20.0 on Amavis (there were some addresses with -3.0, from default config on Ubuntu), does that make it possible that SpamAssassin will mark some emails with a

Re: spamd[18549]: config: failed to parse line, skipping, in /etc/mail/spamassassin/local.cf: use_auto_whitelist 1

2010-05-10 Thread Bowie Bailey
ram wrote: On Wed, May 5, 2010 at 6:49 PM, Bowie Bailey bowie_bai...@buc.com mailto:bowie_bai...@buc.com wrote: ram wrote: i still see this errors May 5 10:28:03.484 [3153] dbg: config: warning: score set for non-existent rule SHORTCIRCUIT May 5

Re: Setting negative pontuation can cause false positives?

2010-05-10 Thread Bowie Bailey
Lauro Costa G. Borges wrote: I configured some addresses to be -9.0 or even -20.0 on Amavis (there were some addresses with -3.0, from default config on Ubuntu), does that make it possible that SpamAssassin will mark some emails with a pontuation higher than they should be, or were in the

Re: yahoo X-YMail-OSG

2010-05-10 Thread J.D. Falk
On May 8, 2010, at 8:18 AM, Michael Scheidell wrote: Yahoo's own DKIM implementation? header signing? Neither. It's related to their anti-spam system, but not intended for end-user parsing. would adding 1 point for each 1K of header length help? Interesting idea! I don't know the precise

Re: Setting negative pontuation can cause false positives?

2010-05-10 Thread Karsten Bräckelmann
No. It is logically impossible, that an additional negative score *causes* the FPs. On Mon, 2010-05-10 at 21:31 +0800, Lauro Costa G. Borges wrote: I configured some addresses to be -9.0 or even -20.0 on Amavis (there were some addresses with -3.0, from default config on Ubuntu), does Err,