> -----Original Message----- > From: shrike-list-admin@xxxxxxxxxx > [mailto:shrike-list-admin@xxxxxxxxxx] On Behalf Of Aaron Konstam > Sent: Tuesday, December 16, 2003 4:42 PM > To: shrike-list@xxxxxxxxxx > Subject: Re: Spam > > > On Tue, Dec 16, 2003 at 11:57:47AM -0600, Jonathan Horne wrote: > > For that kind of SLA, I have a feeling your going to have > to pay some > > money. At my office were deeply in bed with NAI, who owns > the source > > for spamassassin, but with the numbers they say (and > demonstrate) of > > spam that will get thru (which is 500 for every million > messages), we > > laid out over 20grand for a product called brightmail, > which they tout > > 1 in 1million spam messages will get thru. Brightmail > doesn't work by > > rules and scoring of messages, but it keeps a database of > actual real > > spam messages, and updates all its brightmail servers with new spam > > definitions (much the same way virus software works) every > 5 minutes. > > > > So far its working great for us, but in a "free of cost" > environment, > > in the case of spam, I think you might have to be prepared for less > > than perfect results. > > > > I personally use spamassassin on my sendmail server, and so > far it has > > caught ALL the spam messages that have come in, and it has > erroneously > > tagged 2 mails from a friend, but they were just forwards > that I would > > not have read anyway (these 2 in particular were filled > with a ton of > > html tags). > > > > jonathan > As has been said by others spamassassin >= version 1.5 can > be trained and very quickly spam disappears. You can decide > how much spam you want to see by setting the cut off point. > 5-5.5 works well after some training. > -- > ------------------------------------------- > Aaron Konstam > Computer Science > Trinity University > 715 Stadium Dr. > San Antonio, TX 78212-7200 > > telephone: (210)-999-7484 > email:akonstam@xxxxxxxxxxx > > > -- > Shrike-list mailing list > Shrike-list@xxxxxxxxxx > https://www.redhat.com/mailman/listinfo/shrike> -list > What do you have to do to train it? Use sa-learn? I have auto learn enabled in the mailscanner config. That has done a pretty good job. Just need to help it out a bit I guess. No one answered me how to help it learn, something like sa-learn --dbpath /var/spool/spamassassin/ --mbox --spam filename or --ham for mismarked? I haven't tried it yet, but I think this would be correct. Thanks, Jake McHenry Nittany Travel MIS Coordinator http://www.nittanytravel.com -- Shrike-list mailing list Shrike-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/shrike-list
