• Can't get Bayesian Working

    By Andy Page 2 decades ago

    Hi,



    1) Installed nspam.dll into the Domino Program directory

    2) Changed NOTES.INI to load spam in the Extmgr line

    3) Configured the Databases and Signed them using the Servers ID

    4) Create a configuration document with PREP as the type of Bayesian filter.



    However after an hour I'm not getting any emails in mailgood.nsf



    SMTP.kSpam.Bayesian.Action = Copy & Deny

    SMTP.kSpam.Bayesian.Boundary = 90

    SMTP.kSpam.Bayesian.Filter = On

    SMTP.kSpam.Bayesian.Mark = On

    SMTP.kSpam.Bayesian.Prep = On

    SMTP.kSpam.Bayesian.Prep_2 = Off

    SMTP.kSpam.Bayesian.Ratio = 10

    SMTP.kSpam.Conf.AddRecipientsItem = Off

    SMTP.kSpam.Conf.Debug = Off

    SMTP.kSpam.Conf.DefaultAction = Accept

    SMTP.kSpam.Conf.DefaultProbIncrease = 0

    SMTP.kSpam.Conf.FilterFromInt = Off

    SMTP.kSpam.Conf.IID = Suthco

    SMTP.kSpam.Conf.Mark = On

    SMTP.kSpam.Conf.MaxNumbersInFrom = 0

    SMTP.kSpam.Conf.MinFromLength = 0

    SMTP.kSpam.Conf.Reload = On

    SMTP.kSpam.LastMinorError = 01/03/2005 19:44:57: ERROR 5.2.1: kSpam.bload calculating probabilties or not loaded.

    SMTP.kSpam.Version = 1.40 beta 0

    19 statistics found







    Am I missing a step ?????



    Many thanks in advance…



    Andy

    • I'll reply to your email

      By Tom Lyne 2 decades ago

      -tom

    • Same Problem here, ERROR 5.2.1

      By Daniel Stelter 2 decades ago

      Hi Tom & all,



      I have the same problem here.

      SMTP.kSpam.LastMinorError = 15.03.2005 22:22:45: ERROR 5.2.1: kSpam.bload calculating probabilties or not loaded.



      I'm having (Dummy-)Rules and Bayesian is running at ".Prep"

      Looks like it's working fine so far (mails in mailspam and mailgood).

      Just curious whether it's really a "MinorError" or will affect me later running without "Prep"?! ;-)



      Danke! -Daniel.

      • Re:

        By Tom Lyne 2 decades ago

        The 5.2.1 error can be ignored if bload is calculating probabilities at the time the error was posted,



        -tom