On Thu, Jan 15, 2004 at 08:22:37AM +1300, Sidney Markowitz wrote: > Duncan Findlay wrote: > >I'm not really sure why these mails seem to not get hit > > I started running 2.61 on my real mail feed just about a week ago, > initially training Bayes on the most recent 2000 spam and about 500 ham > that I had received. I also have been learning every piece of mail that > comes in as either spam or ham. Since then it has scored BAYES_99 on > every one of the "poisoned" spam with random words after the /HTML tag. > > Could there be something about the way you train or use Bayes that is > causing it to miss on these "poisoned" spam? Why would we see such > different results?
Hmmm... I just checked my Bayes DB and noticed something very very strange. 0.000 0 2 0 non-token data: bayes db version 0.000 0 172 0 non-token data: nspam 0.000 0 407 0 non-token data: nham 0.000 0 21880 0 non-token data: ntokens 0.000 0 1073413315 0 non-token data: oldest atime 0.000 0 1074099954 0 non-token data: newest atime 0.000 0 1073413403 0 non-token data: last journal sync atime 0.000 0 1073413403 0 non-token data: last expiry atime 0.000 0 0 0 non-token data: last expire atime delta 0.000 0 0 0 non-token data: last expire reduction count 0.500 0 0 0 Fog 0.500 0 0 0 anti 0.500 0 0 0 largest 0.500 0 0 0 ile 0.500 0 0 0 requiring 0.500 0 0 0 HX-Spam-Checker:sk:spamass 0.500 0 0 0 Via-gr-a 0.500 0 0 0 spending 0.500 0 0 0 T.V 0.500 0 0 0 System! 0.500 0 0 0 init.d .... So essentially, I have no Bayes DB. I don't know why the counts are 0... that certainly shouldn't happen, should it? Also, I don't know what happened to my cron jobs -- I definitely should have more mail in my DB. I wonder if my DB is corrupt for some reason. I sometimes worry about this when I downgrade spamassassin every once in a while for testing, etc. Anyone seen this before? -- Duncan Findlay
signature.asc
Description: Digital signature
