MS % rate of catch

[Billy A. Pumphrey]

I did some checking.  I searched for the message ID in the maillog for a
email address the I had set to REJECT and yes, the header said someone
else besides what the maillog said, which I wasn't doubting it a bit.

I also did a search for the rule hits that you were talking about,
" rule hits named RCVD_IN_SORBS, RCVD_IN_OPM," and it came up with some,
so I guess that its using DNSBLs.

Here it is:

BIZ_TLD 0.78, DNS_FROM_RFCI_DSN 1.39, NO_REAL_NAME 0
.28, RCVD_IN_BL_SPAMCOP_NET 2.25, RCVD_IN_DSBL 1.10, RCVD_IN_DYNABLOCK
2.55, RCV
D_IN_SORBS 0.10)


-----Original Message-----
From: MailScanner mailing list [mailto:MAILSCANNER at JISCMAIL.AC.UK] On
Behalf Of Billy A. Pumphrey
Sent: Monday, April 05, 2004 5:51 PM
To: MAILSCANNER at JISCMAIL.AC.UK
Subject: Re: MS % rate of catch

Thank you for the info, I'll have to save that for another day, I know
windows but not enough Linux to make my way around very well.
Eventually I want to set both of those up as It seems that they are a
standard to be installed and work well and better with them.  Thank you
for your info.  

PS. If there is a page with instructions on installing and configuring
these, please let me know.

-----Original Message-----
From: MailScanner mailing list [mailto:MAILSCANNER at JISCMAIL.AC.UK] On
Behalf Of Matt Kettler
Sent: Monday, April 05, 2004 5:29 PM
To: MAILSCANNER at JISCMAIL.AC.UK
Subject: Re: MS % rate of catch

At 06:20 PM 4/5/2004, Billy A. Pumphrey wrote:
>Most likely not DNSBLs because I don't know what that is.  I'm guessing
>dns black list where it goes to web sites and checks to see if its
spam,
>and not bayes.  I see all these messages come through the list about
>bayes but I still don't understand it.
>
>Is this DNSBLs in the mailscanner.conf?
>Spam List = ORDB-RBL spamcop.net .... and so on.

Well that is a way of calling a DNSBL (DNS blacklist), but I was
speaking
of DNSBLs in SA.. you need to install the Net::DNS perl module for them
to
go. They show up as rule hits named RCVD_IN_SORBS, RCVD_IN_OPM, etc.

As for bayes, it's a trained statistical analysis system. You feed it
training using sa-learn, and it latter applies those lessons to inbound
email. shows up as rule hits such as BAYES_10 and BAYES_99. Requires the
DB_File perl module in SA 2.6*.