URIBL_SBL timeouts running MS in debug mode.

[donald.dawson at bakerbotts.com]

Excellent information!

-----Original Message-----
From: mailscanner-bounces at lists.mailscanner.info
[mailto:mailscanner-bounces at lists.mailscanner.info] On Behalf Of Gareth
Sent: Friday, November 23, 2007 1:43 PM
To: MailScanner discussion
Subject: RE: URIBL_SBL timeouts running MS in debug mode.


>From logwatch for yesterday:-

 Messages rejected using Anti-Spam site 3799 Time(s)
     zen.spamhaus.org identified 3749 spam messages.
     bl.spamcop.net identified 48 spam messages.
     autoblock.dnsbl identified 2 spam messages.

Messages are rejected in the order shown above so the 48 spamcop blocked
is
what spamhaus missed.

--------------------------------------------------------

>From August when we started testing :-

 Messages rejected using Anti-Spam site 7187 Time(s)
     zen.dnsbl identified 6781 spam messages.
     bl.spamcop.net identified 392 spam messages.
     combined.njabl.org identified 14 spam messages.

We then switched spamcop so that it was first to see how well it did and
how
many zen caught that it missed :-

 Messages rejected using Anti-Spam site 7043 Time(s)
     bl.spamcop.net identified 2654 spam messages.
     zen.dnsbl identified 4385 spam messages.
     combined.njabl.org identified 2 spam messages.

Therefore from our testing we found that of all the rbl detected spam,
spamcop detected 38% while spamhaus detected 95%.
Factoring in the typical 500 spam mesages that get past RBLs these
percentages of detected spam fall to 31% for spamcop and 88% for
spamhaus.

Thats the justification we used for then subscribing to the datafeed
services (tests were performed using a trial account).

> -----Original Message-----
> From: mailscanner-bounces at lists.mailscanner.info
> [mailto:mailscanner-bounces at lists.mailscanner.info]On Behalf Of Steve
> Freegard
> Sent: 23 November 2007 08:39

>
> Add Spamcop to that list (bl.spamcop.net) - it's much more reliable
than
> it used to be, so it can be safely used at the MTA level now.  It is
> almost as effective as the SBL+XBL portions of Spamhaus from my
> own testing.
>
> Remember you can still query the CBL (cbl.abuseat.org) which forms the
> largest and most frequently changed part of the Zen dataset providing
> you fall under their usage terms as well.

-- 
MailScanner mailing list
mailscanner at lists.mailscanner.info
http://lists.mailscanner.info/mailman/listinfo/mailscanner

Before posting, read http://wiki.mailscanner.info/posting

Support MailScanner development - buy the book off the website!