On Thu, 27 Jun 2002, Kevin G. J. Freels wrote: > The whitelist_to section of the doc: > > whitelist_to [EMAIL PROTECTED] > If the given address appears in the `To:' or `Cc:' > headers, mail will be whitelisted. > > As I read this, it says that the whitelist_from is for external > addresses which send mail to us, while whitelist_to is for > internal users (read: us) who send mail out to those sites. > > Is this an incorrect interpretation?
It's incorrect, but it's unrelated to your problem. The SA docs are written as if the only mail being filtered is what is being sent to your site, not being sent from your site. It's a side- effect of using sendmail's milter that your outgoing mail is also getting processed. So whitelist_to (and more_spam_to and all_spam_to) are for recipients who want to bump up the spamminess threshold for messages address directly to themselves, without changing it for messages that are sent to some other address (a forged one, or a mailing list, etc.). Using whitelist_to to mitigate the effects of unintentional outbound SA tagging is an interesting idea, but not the intended usage. > [...] whitelist is applied, bumping the score down to change > is as non-spam-status. But it is still getting mangled by > mime_defang to a somewhat-unreadable form (straight HTML code). Sounds like a MIMEDefang problem. Does it happen only when the messages pass through SA? That is, if you "unplug" SA from MIMEDefang but leave MIMEDefang itself in place, what happens? ------------------------------------------------------- Sponsored by: ThinkGeek at http://www.ThinkGeek.com/ _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
