> Richard Smits wrote: > >Hos safe is it to pump up the score for the ANY_BOUNCE_MESSAGE ? > >Is it bug free, so I can give it 5 or 10 points ?
On 18.04.08 09:19, Jason Haar wrote: > So you are wanting to mark ANY bounce, out of office, or mailing-list > related email into your organization as spam? If you want to do that, > then sure! :-) > > My own investigations would show that would not be a good idea. I think > you meant "BOUNCE_MESSAGE" instead - but even that is catching stuff > that isn't backscatter. yes, since (according to previous discussion) VBounce was not designed to mark backscatter as spam, but to mark (suspicious) bounces as bounces. It probably needs many changed to be reliable in the way most users expect - to catch baskscatter while not catch other > ...and I don't think the Backscatter FAQ answers this question. IMHO > VBounce tags *bounces* - not backscatter. Backscatter is a *subset* of > bounces - so it tags stuff that isn't backscatter. whitelist_bounce_relays should whitelist non-backscatter bounces, but that might not be enough. For example, I wonder why does not VBounce look at Received: headers to see if it came from hosts in internal network, such bounces will surely not be backscatter imho. > I'm working on a backscatter.cf to exclusively catch backscatter - but > it's still tagging incorrect stuff. (all my Sourceforge moderator mail > for starters). If I get it working reliably, I'll flick it up the food > chain... good luck. -- Matus UHLAR - fantomas, [EMAIL PROTECTED] ; http://www.fantomas.sk/ Warning: I wish NOT to receive e-mail advertising to this address. Varovanie: na tuto adresu chcem NEDOSTAVAT akukolvek reklamnu postu. Emacs is a complicated operating system without good text editor.
