> > Just curious as to what average percent of spam people see > SURBL hitting. In a non scientific manor, I average about 85% > or greater hitting SURBL for all spam that doesn't get > rejected by my MTA. I have a feeling if I clean up my results > a bit, that number would be even higher. >
Okay, let me give you some real numbers from our inhouse mail for the last week.. As you can see, half the top rules are network tests, so people running without them (-L) are really missing out.. TOP SPAM RULES FIRED ------------------------------------------------------------------------ RANK RULE NAME COUNT %OFRULES %OFMAIL %OFSPAM ------------------------------------------------------------------------ 1 URIBL_SBL 1739 6.66% 50.71% 85.04% 2 HTML_MESSAGE 1586 6.07% 46.25% 77.56% 3 URIBL_WS_SURBL 1447 5.54% 42.20% 70.76% 4 URIBL_JP_SURBL 1373 5.26% 40.04% 67.14% 5 URIBL_OB_SURBL 1364 5.22% 39.78% 66.70% 6 DCC_CHECK 1047 4.01% 30.53% 51.20% 7 MIME_HTML_ONLY 995 3.81% 29.02% 48.66% 8 BAYES_99 949 3.63% 27.68% 46.41% 9 PYZOR_CHECK 860 3.29% 25.08% 42.05% 10 DIGEST_MULTIPLE 836 3.20% 24.38% 40.88% 11 URIBL_SC_SURBL 635 2.43% 18.52% 31.05% 12 RAZOR2_CF_RANGE_51_100 612 2.34% 17.85% 29.93% 13 RAZOR2_CHECK 604 2.31% 17.61% 29.54% 14 HTML_WEB_BUGS 496 1.90% 14.46% 24.25% 15 HTML_90_100 431 1.65% 12.57% 21.08% 16 HTML_IMAGE_RATIO_02 426 1.63% 12.42% 20.83% 17 MSGID_SPAM_LETTERS 330 1.26% 9.62% 16.14% 18 HTML_LINK_IMAGE_BUG 308 1.18% 8.98% 15.06% 19 URIBL_AB_SURBL 292 1.12% 8.52% 14.28% 20 BAYES_50 278 1.06% 8.11% 13.59% ------------------------------------------------------------------------ Oh and this is out of the stuff that actually makes it to the queue... Which is normally like 8-15% of the overall total connections, so these %OFSPAM numbers are probably conservative, but I really don't feel like shutting down rbl's just to get a true number :) d
