Nigel Frankcom wrote:
Is it necessary to remove SA headers during the learn? I use two
methods arbitrarily;
From man sa-learn (SA 2.64):
If the messages you are learning from have already been filtered through
SpamAssassin, the learner will compensate for this. In effect, it
learns what each message would look like if you had run "spamassassin
-d" over it in advance.
From man sa-learn (SA 3.1.7):
If the messages you are learning from have already been filtered through
SpamAssassin, the learner will compensate for this. In effect, it
learns what each message would look like if you had run "spamassassin
-d" over it in advance.
<g>
IOW, sa-learn "knows" what headers SpamAssassin adds.
I should point out that it does NOT inherently understand headers added
by other scanning systems - including those that either call the
SpamAssassin Perl libraries (Amavis, MIMEDefang), or call spamassassin
or spamc (qmail-<something>) but add their own headers based on the
results of the scan. However, bayes_ignore_header <header name> in your
main SA config should compensate for that.
-kgd
