I have few users which if the email is spam it has to be delivered to theirs mailboxes.
I used "whitelist_to" parametr but there are some meassages which are blocked.
From docs:
There are three levels of To-whitelisting, "whitelist_to", "more_spam_to" and "all_spam_to". Users in the first level may still get some spammish mails blocked, but users in "all_spam_to" should never get mail blocked.
I would like to know if the string "... should never get mail blocked" is true :-)
all_spam_to provides a -100 point score. That's a pretty hefty nonspam bias, and unless you've been jacking spam rules up into the +30 range, it should be effective.
However, beware... SA cannot always determine who the recipient of a message is. It does not get a copy of the envelope, thus it must try to decipher the recipient from the headers alone. If the message is Bcc'ed and your MTA doesn't insert a "for [EMAIL PROTECTED]" in the received headers, SA will not know who the message is being sent to, and all_spam_to will fail.
In general, absolute whitelists are generaly best done by going around SA in the tool that calls SA.. ie: using procmail rules to skip the call. You save CPU time this way too....
