Markus Braun wrote: > > I have made > > spamassassin -D --lint > > but i got the warnings in the attached file.
>> warning: description for EXCUSE_6 is over 50 chars This is fairly common. It indicates that your SA and ruleset versions do not match. Search the mailing list logs. This problem comes up regularly. Once you get this taken care of, you will probably have more rules to work with and, therefore, higher scores in general. >> debug: is Net::DNS::Resolver available? no >> debug: is DNS available? 0 You need to install (or update) the perl module Net::DNS. Once you get this installed the network tests will start working. They provide quite a bit of help in catching spam. >> debug: tests=ALL_TRUSTED,MISSING_DATE,MISSING_SUBJECT,NO_REAL_NAME The presence of ALL_TRUSTED in an external email (spam or not) is an indication that your trust path is messed up. You need to configure trusted_networks in your local.cf. Read the wiki on this here: http://wiki.apache.org/spamassassin/TrustPath >> debug: Razor2 is not available >> debug: Pyzor is not available: pyzor not found >> debug: DCC is not available: no executable dccproc found. Razor2, Pyzor, and DCC can be a big help in finding spam. Razor2 in particular is very easy to install and configure. Once you get the rest of these issues taken care of, you may want to look into it. > But this is the same in the last email, which i have made with > sa--learn to spam. >> debug: bayes: Not available for scanning, only 1 spam(s) in Bayes DB < 200 You haven't learned 200 spam messages yet, so Bayes will not attempt to score the message. In any case, learning one message as spam does not guarantee that the next identical message will be marked with BAYES_99. Every time you learn a message as spam, the score will go up for messages containing similar words. You just have to keep training it and be patient. -- Bowie
