Nathan England wrote: > Anyone have any good spamassassin tricks while I'm on the topic? > I have mine set to change the header of anything over a 4.0 but it is very > rare that anything is given over a 4.0. Even spam will only get a few points > over a 4.0 > > What are people doing to get a 20.0 as some tutorials say to set it at? > If I don't have network tests working then I will get low scores like yours. Make sure you /etc/host file is correct because spamassassin is funky about that and auto enableing network tests. You should see the bold stuff showing up in spam. Otherwise do the math and you'd see this wouldn't get a very high score. pts rule name description ---- ---------------------- -------------------------------------------------- 1.7 DATE_IN_FUTURE_06_12 Date: is 6 to 12 hours after Received: date 1.3 INFO_TLD URI: Contains an URL in the INFO top-level domain 4.0 BAYES_99 BODY: Bayesian spam probability is 99 to 100% [score: 1.0000] *0.2 DNS_FROM_RFC_ABUSE RBL: Envelope sender in abuse.rfc-ignorant.org 1.4 DNS_FROM_RFC_WHOIS RBL: Envelope sender in whois.rfc-ignorant.org 2.6 DNS_FROM_RFC_DSN RBL: Envelope sender in dsn.rfc-ignorant.org 2.0 RCVD_IN_SORBS_DUL RBL: SORBS: sent directly from dynamic IP address [155.143.235.19 listed in dnsbl.sorbs.net] 2.2 RCVD_IN_WHOIS_INVALID RBL: CompleteWhois: sender on invalid IP block [155.143.235.19 listed in combined-HIB.dnsiplists.completewhois.com] 1.7 DNS_FROM_RFC_POST RBL: Envelope sender in postmaster.rfc-ignorant.org 0.2 DNS_FROM_AHBL_RHSBL RBL: From: sender listed in dnsbl.ahbl.org 1.6 URIBL_SBL Contains an URL listed in the SBL blocklist [URIs: hostelcontact.info] 4.1 URIBL_JP_SURBL Contains an URL listed in the JP SURBL blocklist [URIs: hostelcontact.info] 3.0 URIBL_OB_SURBL Contains an URL listed in the OB SURBL blocklist [URIs: hostelcontact.info]* 2.6 REPTO_OVERQUOTE_THEBAT The Bat! doesn't do quoting like this The other bulk of my scores come from the baysian engine... Once you have 200 hams and 200 spams that turns on automatically. (read about sa-learn for that one) To know how many you have or how many you need to go, do a "sa-learn --dump magic". Notice on my spam that is this score 4.0 BAYES_99 BODY: Bayesian spam probability is 99 to 100% [score: 1.0000] Normally I adjust BAYES_99 to 4.5 points but since I just upgraded spamassassin I have everything at default to see how the detection/rules have changed before I start modifying things. For me, with thousands of hams and thousands of spams, this thing is literally never wrong all by itself for me. Thats why I let it mark spam almost all by itself. But results vary depending on everybody, and what you've taught it. Anywho, tally up the other 3 failed on that email and thats how a lot of things would be if both of these things weren't working. Sound familiar? -Mike --------------------------------------------------- PLUG-discuss mailing list - PLUG-discuss@lists.plug.phoenix.az.us To subscribe, unsubscribe, or to change you mail settings: http://lists.PLUG.phoenix.az.us/mailman/listinfo/plug-discuss