Salvatore Toribio ([EMAIL PROTECTED]) wrote: > > At 15:22 +0000 22-03-2004, Assured Computing wrote: > >Hi, > > > >I have a couple of installs where SpamAssassin is happily filtering mail more > >or less correctly however I noticed that it does not seem to be autolearning. > >How do I make it autolearn? Here is one of the headers: > >X-Spam-Status: Yes, hits=31.0 required=5.0 > >tests=BANG_EXERCISE,BANG_GUARANTEE,ntCLICK_BELOW_CAPS,DATE_IN_FUTURE_03_06,DATE_SPAMWARE_Y2K,n > >tFORGED_MUA_OUTLOOK,FORGED_OUTLOOK_TAGS,GUARANTEED_STUFF,HTML_50_60,n > >tHTML_FONTCOLOR_BLUE,HTML_FONTCOLOR_GREEN,HTML_FONTCOLOR_RED,ntHTML_FONT_BIG,HTML_IMAGE_ONLY_08,HTML_LINK_CLICK_CAPS,n > >tHTML_LINK_CLICK_HERE,HTML_MESSAGE,HTML_MIME_NO_HTML_TAG,IMPOTENCE,n > >tMIME_HTML_NO_CHARSET,MIME_HTML_ONLY,MIME_HTML_ONLY_MULTI,ntMISSING_MIMEOLE,MONEY_BACK,PENIS_ENLARGE,PENIS_ENLARGE2 > >autolearn=no ntversion=2.63 > > > >Why doesn't autolearn=yes? > > > >This is a fresh install using the defaults. > >Slackware 9.1 > >SpamAssassin installed through Perl. > >Summary of my perl5 (revision 5.0 version 8 subversion 3) configuration: > > > > Don't worry about that. I wondered about the same behavior and I > found the explanation somewhere in spamassassin docs. I've the debug > mode of spamassassin on in one server look below > > This messages was autolearned: > > spamd[23566]: debug: auto-learn? ham=-0.2, spam=15.7, > body-hits=9.125, head-hits=29.581 > spamd[23566]: debug: auto-learn: currently using scoreset 3. > recomputing score based on scoreset 1. > spamd[23566]: debug: auto-learn: original score: 30.782, recomputed > score: 28.311 > spamd[23566]: debug: auto-learn? yes, spam (28.311 > 15.7) > > > > And this one not, even its score was 19.067 greater than 15.7: > > spamd[24686]: debug: auto-learn? ham=-0.2, spam=15.7, > body-hits=7.958, head-hits=13.576 > spamd[24686]: debug: auto-learn: currently using scoreset 3. > recomputing score based on scoreset 1. > spamd[24686]: debug: auto-learn: original score: 19.067, recomputed > score: 9.487 > spamd[24686]: debug: auto-learn? no: inside auto-learn thresholds > > > I found that body-hits and head-hits must be greater than some value, > otherwise the message it is not autolearned. Search in the > spamassassin docs. > > Cheers > > Salvatore > If I remember right, the score has to be over 12 to trigger autolearn.
I'll stop worrying thanks. ------------------------------------------------------- This SF.Net email is sponsored by: IBM Linux Tutorials Free Linux tutorial presented by Daniel Robbins, President and CEO of GenToo technologies. Learn everything from fundamentals to system administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click _______________________________________________ Qmail-scanner-general mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/qmail-scanner-general
