On December 18, 2016, Karen Lewellen wrote: > We have Spam assassin here, and I do run sa-learn. > However, I do not run this on addresses that use real names, even > if I know the material is spam. > At what level do programs like spam assassin do the filtering? Are > there other steps I can take to curb the flow? SpamAssassin (SA) takes the content into consideration, so even if you're receiving spam from known-good addresses, feed it to SA. It will learn from the content, not just the sender. Depending on how your mail-server is configured, you can also set up catch-all accounts and then direct various sources to custom email addresses. Because I own our domain, I can have all mail to this list come to blinux.list@<mydomain> which lets me set up filters to the effect "if anything comes to blinux.list@ and isn't also either to the mailing-list or in my personal address-book of people with whom I've previously sent messages, treat it as junk" You can do something similar with GMail which allows you to take your username@xxxxxxxxx and append a plus-sign and a tag for filtering such as "username+blinux@xxxxxxxxx". Not knowing the peculiars of your particular mail configuration, it's hard to offer better suggestions, but at least you now know enough to train SA with the junk mails even if they come from known-good addresses. -tim _______________________________________________ Blinux-list mailing list Blinux-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/blinux-list