Open shawniverson opened 3 years ago
Testing postfix changes internally in dev env.
Just a thought:
Adding rbls like "reject_rbl_client zen.spamhaus.org, reject_rbl_client bl.spamcop.net" to the smtpd_recipient_restrictions may not be desirable and should be optional with a suitable explanation.
My objection is that false positives would be rejected without us ever learning about it, potentially causing a problem in communications.
Personally, I prefer using the RBS at the spamassassin level. If a false positive is marked as high spam, at least I have a chance of catching and releasing it as I'll have it stored in my system. Not so if we reject it at the postfix level.
(that said, spamhaus and spamcop tend to have few false positives)
https://forum.efa-project.org/viewtopic.php?f=14&t=4477