Hello Jonathan,
Tuesday, January 20, 2004, 9:45:23 AM, you wrote:
JN> Time to feed Bayes again.. I think I have almost 1,000 spams in my
JN> spam folder (I feed it when it hits 1000)
Why do you wait? I feed Bayes at least once a day, sometimes two or
three times.
True, I get 700-800 spam each
On Tue, 2004-01-20 at 16:12, Jonathan Nichols wrote:
> http://www.pbp.net/~jnichols/spam.txt
>
> It also slipped right by Mailscanner on another host, but I'm surprised
> that it scored 0.0 on my SA setup (backhair, weeds, everything in
> rules_du_jour)
FWIW, here's what it scored on my system:
5.4 BAYES_99 BODY: Bayesian spam probability is 99 to 100%
[score: 1.]
1.1 RAZOR2_CF_RANGE_51_100 BODY: Razor2 gives confidence between 51 and 100
[cf: 100]
1.0 RAZOR2_CHECK Listed in Razor2 (http://razor.sf.net
Jonathan
Worth changing the X-MailScanner header stuff to be
X-%org-name%-Mailscanner in the MailScanner.conf. This was introduced at
MS 4.24 (I think) as one of the virus's at the time was putting this in
the header to stop MS hosts virus scanning the email!
The other headers in there look '
On Tue, 20 Jan 2004, Jonathan Nichols wrote:
> http://www.pbp.net/~jnichols/spam.txt
That's really odd, here it tripped a DATE_IN_PAST rule. Here's the report:
Content analysis details: (12.4 points, 5.0 required)
pts rule name description
--
--
Jonathan Nichols wrote:
http://www.pbp.net/~jnichols/spam.txt
It also slipped right by Mailscanner on another host, but I'm surprised
that it scored 0.0 on my SA setup (backhair, weeds, everything in
rules_du_jour)
---
The SF.Net email is sp
http://www.pbp.net/~jnichols/spam.txt
It also slipped right by Mailscanner on another host, but I'm surprised
that it scored 0.0 on my SA setup (backhair, weeds, everything in
rules_du_jour)
---
The SF.Net email is sponsored by EclipseCon 20
I haven't seen this before, and it appears particularly nasty.
X-Spam-Status: No, hits=3.6 required=5.0
tests=BASE64_ENC_TEXT,HTML_60_70,HTML_FONT_BIG,
HTML_FONT_COLOR_RED,MIME_HTML_ONLY,MIME_MISSING_BOUNDARY
version=2.55
X-Spam-Level: ***
X-Spam-Checker-Version: SpamAssassin
Craig Hughes
Cc: [EMAIL PROTECTED]
Subject: Re: [SAtalk] One That Got Through
On 26 Feb 2002, Craig Hughes wrote:
> Yeah, increasing discrimination will end up causing the scores for
> some false-negatives to decrease. But hey, quit complaining!
I'm not complaining; I'm just offe
On 26 Feb 2002, Craig Hughes wrote:
> Yeah, increasing discrimination will end up causing the scores for some
> false-negatives to decrease. But hey, quit complaining!
I'm not complaining; I'm just offering examples in case somebody wants to
concoct new rules, which I don't have time for, just
Yeah, increasing discrimination will end up causing the scores for some
false-negatives to decrease. But hey, quit complaining! Your
false-negative rate has been reduced to 1/3 of its previous level!
C
On Tue, 2002-02-26 at 12:03, Bart Schaefer wrote:
> On 26 Feb 2002, Craig Hughes wrote:
>
>
On 26 Feb 2002, Craig Hughes wrote:
> Latest CVS (soon to be 2.1) catches the viagra, and also the unsubscribe
> details.
The other day I reported that I had about an 88% success rate on some
recent spam with SA 2.01. Actually it was a bit lower than that; it
missed 292 messages out of 1926.
Latest CVS (soon to be 2.1) catches the viagra, and also the unsubscribe
details.
C
On Tue, 2002-02-26 at 11:19, Erik B. Berry wrote:
> It seems the SA 2.01 rules missed the unique ID in the subject and body
> here, as well as the removal instructions. Maybe this points to
> possibilities for
It seems the SA 2.01 rules missed the unique ID in the subject and body
here, as well as the removal instructions. Maybe this points to
possibilities for spam phrases for things like "viagra pharmacy",
"no hassle", "reasonable prices", "fast delivery", "check out the site",
"highest quality"?
14 matches
Mail list logo
