I'm learning to understand how to properly set up a site-wide bayes
database on my server. Thanks for everyone's help and patience so far.
I've discovered that the SA score assigned to a user's incoming email is
different than the SA score run through the "spamc" or "spamassassin"
command. For example, the SA headers for email "A" will show a score of
only 1.4 (non-spam) in the user's inbox. It shows as non-span despite
the fact that I have run it through sa-learn as spam. When I run the
same email through "spamc -R < <message_file>" on the command line as
the same user that received the original message, I see a score of 6.8
and it is properly getting classified as spam.
I'm trying to determine what accounts for the different scores and fix
this problem so the correct score is assigned to mail coming into the
user's inbox.
After doing some investigating, I discovered the user still had a
.spamassassin directory in their home directory. The directory has only
a single "user_prefs" file. But I'm wondering if the existence of this
directory might cause spamassassin filter to ignore site-wide bayes
database. If that's not the problem, what might account for the
different scores and how might I fix the issue?
- Can a .spamassassin directory in a user's home dire... Steve Dondley
-
