Theo Van Dinter wrote:
On Thu, Jun 01, 2006 at 01:41:36PM -0400, DAve wrote:
Currently 3.0.4 on the toasters, 3.0.2 on the MailScanner boxes. These
[...]
http://pixelhammer.com/spam/spam1.txt
http://pixelhammer.com/spam/spam2.txt
http://pixelhammer.com/spam/spam3.txt
http://pixelhammer.com/spam/spam4.txt
http://pixelhammer.com/spam/spam5.txt
http://pixelhammer.com/spam/spam6.txt
FWIW, running 3.1.2 w/ sa-update these are caught without issue:
[7363] dbg: check: is spam? score=12.342 required=5
[7363] dbg: check:
tests=BAYES_99,HTML_30_40,HTML_MESSAGE,RCVD_IN_BL_SPAMCOP_NET,RCVD_IN_XBL,SPF_NEUTRAL,UNPARSEABLE_RELAY
[7376] dbg: check: is spam? score=12.093 required=5
[7376] dbg: check:
tests=BAYES_50,DATE_IN_PAST_48_96,DCC_CHECK,DIGEST_MULTIPLE,DNS_FROM_RFC_ABUSE,FORGED_RCVD_HELO,RAZOR2_CF_RANGE_51_100,RAZOR2_CF_RANGE_E4_51_100,RAZOR2_CHECK,RCVD_IN_SORBS_DUL,RCVD_IN_XBL
[7381] dbg: check: is spam? score=11.622 required=5
[7381] dbg: check:
tests=BAYES_99,MISSING_HB_SEP,MISSING_SUBJECT,NORMAL_HTTP_TO_IP,RCVD_IN_BL_SPAMCOP_NET,TO_CC_NONE
[7385] dbg: check: is spam? score=15.448 required=5
[7385] dbg: check:
tests=BAYES_50,INVESTMENT_ADVICE,MISSING_HB_SEP,MISSING_SUBJECT,RCVD_IN_BL_SPAMCOP_NET,TO_CC_NONE,TVD_ACT_193,TVD_FUZZY_SECURITIES
[7389] dbg: check: is spam? score=12.903 required=5
[7389] dbg: check:
tests=ADVANCE_FEE_1,ADVANCE_FEE_2,ADVANCE_FEE_3,ADVANCE_FEE_4,BAYES_50,MISSING_HB_SEP,MISSING_SUBJECT,TO_CC_NONE
[7398] dbg: check: is spam? score=7.449 required=5
[7398] dbg: check:
tests=BAYES_50,MISSING_HB_SEP,MISSING_SUBJECT,TO_CC_NONE,UNPARSEABLE_RELAY,URIBL_BLACK
Theo, I appreciate the results, that means more to me than "upgrade".
Results speak louder than anything else. Could you tell me the scores
for each test? If you are as busy as I am, I understand if you can not
rerun everything. I currently do not run the network tests as I do all
my RBL tests in the MTA at connection time. Razor and DCC, well, maybe
time to try them again. SpamCop, I won't go there, we have issues with
the people at SpamCop. I neither use them nor trust them.
I have both my mailgateways staged for an upgrade of MailScanner and the
addition of MailWatch. I have the setup configured for SA 3.1.2 and
running sa_update, which I think was a much needed addition to SA. Looks
like I should go ahead with the SA upgrade now rather than later.
Bayes, arrgg!! More than once I've been given examples of bayes being
the solution I need. I really really really want bayes to work. But each
time I set it up, the db gets huge, scan times go through the roof, and
I end up disappointed. The one time it worked for me I ended up training
by hand a couple hours a week, then an hour a week of my time, then all
on my own time. The result was I caught 20,000 spam a day instead of
19,500 spam a day.
Right now I am getting the following results,
Email: 294914 Autolearn: 0 AvgScore: 12.06 AvgScanTime: 4.01 sec
Spam: 192276 Autolearn: 0 AvgScore: 19.92 AvgScanTime: 3.21 sec
Ham: 102638 Autolearn: 0 AvgScore: -2.66 AvgScanTime: 5.51 sec
Time Spent Running SA: 328.51 hours
Time Spent Processing Spam: 171.35 hours
Time Spent Processing Ham: 157.16 hours
Possibly I need to use SQL bayes and tie my two MailScanner boxes
together, running one common bayes for all users. I have a dedicated SQL
server for mail and SA, it runs on a private 1gb connection to the mail
servers.
Thoughts?
DAve
--
Three years now I've asked Google why they don't have a
logo change for Memorial Day. Why do they choose to do logos
for other non-international holidays, but nothing for
Veterans?
Maybe they forgot who made that choice possible.
