Hello Andreas,
Am Mittwoch, 31. Oktober 2012, 10:16:06 schrieb Andreas Kusalananda Kähäri:
> You say $USER is set up correctly, but is it really?
> On some systems $USER is automatically set to the "current user", i.e.
> the user running the script. Is this root or you or dspam, or is it
> really the user?
I have virtual domains. The $USER variable in that script contains the
complete email address of the user.
> I also notice you're using $user (below, in lower-case). Is there a
> confusion here?
That was just a shortcut. In fact, I called
dspam_stats <user email>
Thanks
Chris
>
> Cheers,
> K~
>
> > > I even cleared the database and repeated the basic training by feeding
> > > ham
> > > and spam.
> > >
> > > This just has no effect. The output of dspam_stats is:
> > > sudo dspam_stats $user
> > > <user> TP: 0 TN: 0 FP: 0 FN: 0 SC: 0 NC: 0
> > >
> > > These values never change. The user still receives spam from similar
> > > sources. No error messages are found in the log files.
> > >
> > > The mails are tagged like this:
> > > X-DSPAM-Result: Innocent
> > > X-DSPAM-Processed: Tue Oct 30 12:18:13 2012
> > > X-DSPAM-Confidence: 0.8334
> > > X-DSPAM-Probability: 0.0000
> > > X-DSPAM-Signature: 1,508fb775183501549213244
> > > X-DSPAM-Factors: 27, 2012, 0.00359, Received*Oct, 0.00472, Received*Oct,
> > > 0.00472, File, 0.00561, Received*ESMTPA, 0.00604,
> > > Received*mail.xxxxx.de,
> > > 0.00659, Date*2012, 0.00659, Received*2012, 0.00659, Received*2012,
> > > 0.00659, Received*xxxxx.de>, 0.00983, Received*xxx, 0.00983, Euros,
> > > 0.01000, Serial, 0.01000, Lucky, 0.01000, 122, 0.01000, Date*Oct,
> > > 0.01403, season, 0.02487, 93, 0.02703, Received*User, 0.03140, release,
> > > 0.04513, among, 0.04854, emerged, 0.07218,
> > > Content-Type*charset="Windows, 0.91659, sum, 0.91591, 450, 0.11478,
> > > Winners, 0.11478, winning, 0.12535
> > >
> > > Any ideas?
> >
> > More info:
> > dspam_stats -H
> >
> > dspam:
> > TP True Positives: 5497
> > TN True Negatives: 1091
> > FP False Positives: 22
> > FN False Negatives: 102
> > SC Spam Corpusfed: 0
> > NC Nonspam Corpusfed: 0
> > TL Training Left: 1387
> > SHR Spam Hit Rate 98.18%
> > HSR Ham Strike Rate: 1.98%
> > PPV Positive predictive value: 99.60%
> > OCA Overall Accuracy: 98.15%
> >
> > Here, TL Training Left is decreasing with good mail and increasing with
> > retraining. Also, in /var/spool/dspam/system.log I find retraining
> > logs:
> > 1351676037 M "Barr. Lonnox Lux (Q.C)."<vrutg...@aol.com>
> > 1,509065a222497195811560 UPDATE... 0.113932 dspam
> > Retrained <20121030163327.1b550113...@mail.3838hz.net>
> >
> > But my users do not get entries in dspam_stats, and still get spam. Whats
> > wrong?
> > Chris
> >
> > --------------------------------------------------------------------------
> > ---- Everyone hates slow websites. So do we.
> > Make your web apps faster with AppDynamics
> > Download AppDynamics Lite for free today:
> > http://p.sf.net/sfu/appdyn_sfd2d_oct
> >
> > _______________________________________________
> > Dspam-user mailing list
> > Dspam-user@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/dspam-user
------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_sfd2d_oct
_______________________________________________
Dspam-user mailing list
Dspam-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/dspam-user
