On Wed, Apr 18, 2012 at 08:55:11PM +0100, Paul Cockings wrote: > On 18/04/2012 20:45, Steve Fatula wrote: > > > > > > On my setup DSPAM is disabled for most users as they don't need > > it, the checks that are in place at the MTA are sufficient enough > > for those users to receive a very low level of spam. DSPAM is > > used for some troublesome accounts or those that want/can handle > > training a filter. > > > >That's great, but I want it enabled. > > > >So, does no one else have any responses to help me with the hit > >rate? I see people have posted such questions in the past and got > >some good replies. I just don't see any for the hit rate that are > >useful. > > > >I just don't get why my hit rate is so very low, when I see > >numerous people with hit rates way above 90%. It's not proving as > >useful as I used to get years ago out of dspam. Perhaps later > >changes have made it less reliable? > > Well, two people have responded and both say try TOE. I would > change to TOE, clear the database and give it few days, report back. > What is stopping you from trying that advice? >
I will go ahead and chime in for TOE. We are running one of the older versions of DSPAM and TEFT users invariably start to lose accuracy as the imbalance of SPAM mail versus not-SPAM mail takes its toll. The fix is to clear their tokens and use TOE, with very good results. Regards, Ken ------------------------------------------------------------------------------ Better than sec? Nothing is better than sec when it comes to monitoring Big Data applications. Try Boundary one-second resolution app monitoring today. Free. http://p.sf.net/sfu/Boundary-dev2dev _______________________________________________ Dspam-user mailing list Dspam-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dspam-user
