I just moved to a new hosting provider who has Spamassassin 3.2.4 running (on
some kind of Linux, 2.6.18-92.1.22.el5) and I'm otherwise unfamiliar with SA.
I'd like some degree of control over what SA is doing, but config for this is
proving confusing for me.
Ideally if I could get SA just to mark up headers with its observations /
scoring, I'd have a good idea of where to set score threshold while I sort
mails correctly for later training. First monitor, then act. However, SA
currently appears to be making spam judgement and to be bayes autolearning.
(A reasonable default setup from the hosting provider.)
I think these things would help me to achieve my goals:
* How do I determine what the current SA config is? Specifically, can I see
whether bayes is enabled, and whether it's auto-learning (if that's
distinct from merely enabled)? Anyway, seeing the whole config would be
useful. The best I've done so far is `spamassassin -D < /dev/null 2>&1 |
less` and `sa-learn --dump magic`. (I may not actually have read access
to the site-wide SA config file, and `locate` is twitted.)
* Can I stop SA from judging spamminess (that is, making the binary
declaration of whether something is spam, X-Spam-Status, X-Spam-Flag) and
retain the scoring markup? I suppose this may not be important, as
sa-learn is said to ignore prior SA markup, it's just that having the
declaration sitting in the headers from there on makes these mails look
spammy whether they truly are, and other more naive tools might be misled.
* If I can't stop SA from judging spamminess, can I at least override the
site-wide config to mark up subjects? I can't figure this out. Currently
I have 'rewrite_header subject ""', but that fails. The docs say the
string should be set to 'a null value', but the config file's syntax for
specifying nulls is not described. Again, this may not be critical, as
sa-learn might ignore the markup, but seeing as the markup is an arbitrary
string, how would sa-learn figure out to ignore the specific string
'***SPAM***' if the config ever changes?
Thanks,
RSK
