Since you're using bayes do you get the same errors when you try to use
sa-learn?
If this was me I'd simply run sa-learn in the perl debugger and debug
how the bayesstore is being written.
I don't think your problem is spamd specific.
On Dec 19, 2003, at 3:06 AM, Lukreme wrote:
I make sure th
I'm not seeing any problems with 5.8.0 and I'm making heavy use of
SpamAssassin in my
perl filter scripts.
On Dec 22, 2003, at 10:22 AM, J. S. Greenfield wrote:
I've seen a few comment here that SA has issues with perl 5.8.
Anybody have a thumbnail sketch of what the issues are? Looks like
R
I cannot think of any legimate emails that would contain X-MT in their
header so why not
just right the rule to check for that instead of worrying about which
digits are present. Then
you only have to worry about what happens when they change the two
characters after the X- :-)
On Dec 22, 2003,
All of this sounds like where you then have you put your faith in
Bayes. Surely there are terms in these emails that are sent that
Spammers aren't using in theirs.
---
This SF.net email is sponsored by: IBM Linux Tutorials.
Become an expert i
Part of my spam filter checking consists of a known domains list. ie. I
never thru any mail that is from a user from that domain into
SpamAssassin. Likewise my whitelist. This could eventually be abused by
spammers quite easily but so far I haven't found it to be a problem. I
can always resort
Are you making use of any of the distributed clearing houses such as
DCC etc?
On Dec 21, 2003, at 4:05 PM, Barry Callahan wrote:
I installed Spamassassin from a RedHat RPM as a test a day or two ago,
and it's properly flagging about 1/3 of the incoming SPAM as such. I
have not played around w
me as ham?
On Dec 19, 2003, at 11:04 AM, Theo Van Dinter wrote:
On Fri, Dec 19, 2003 at 10:12:25AM -0600, Robert Nicholson wrote:
So how do you forget a message then?
why do you want to forget a message?
#forget old spam
$spamtest->learn($learnmail, undef, 0, 1);
#lear
e nham, hspam counts haven't
changed.
On Dec 19, 2003, at 8:47 AM, Robert Nicholson wrote:
Yes I've seen the comments. Mail::Audit supports Maildir and
implements forward. Two things
I need for my solution.
Again though it's puzzling to me why sometimes I can learn and other
times I
the
Mail::Audit.
On Dec 19, 2003, at 7:53 AM, Theo Van Dinter wrote:
On Fri, Dec 19, 2003 at 04:06:17AM -0600, Robert Nicholson wrote:
Other than this I've been able to use Mail::Audit with Spamassassin
but
is there
another way to go about this then?
We've pretty much already dropped
I'm guessing a little here but it looks to me as though if I have
create a brand new Mail::Audit
and I wish to call learn. The way things are setup at the moment is
that it assumes I've previously called check or something because
otherwise the mail object won't be a NoMailAudit which I think Ba
I'm using Mail::Audit as my mail object but i think I have a problem
with Bayes expecting
there to be a get_pristine_body method on a Mail::Audit
There isn't one.
Other than this I've been able to use Mail::Audit with Spamassassin but
is there
another way to go about this then?
-
the mail as ham but forget what we already recorded
On Dec 13, 2003, at 6:26 PM, Matt Kettler wrote:
At 06:53 PM 12/13/2003, Robert Nicholson wrote:
$status = $f->learn ($mail, $id, $isspam, $forget)
suggests that both $isspam and $forget are significant however
it will only ever do one
MUA easily allowed me to include the
original mail as a MIME attachment but I have to save each message to
disk in order to do that. Way too tedious.
On Dec 18, 2003, at 4:47 PM, David Smith wrote:
On Thu, 2003-12-18 at 16:27, Robert Nicholson wrote:
So I was thinking about how one could maintain Baye
So I was thinking about how one could maintain Bayes from an MUA for
certain messages.
Specifically whenever you see false positives in your spam that perhaps
were learnt by bayes
you could have something described below.
Assuming you can fetch mail with courier IMAP thru it's message-id and
n
Anybody know how to reply or forward but still have the original
message-id in order to forget
an incorrect learnt spam message?
The idea is that my filter is gonna learn spam or ham depending on
status->is_spam
but I just want to reply thru my filter and capture the original
message id and un
I have my own domain and I have rules that automatically kick out any
message not addressed to recognised users at my domain. None of these
messages get anywhere
near SpamAssassin. Those that do I'm finding that required_hits of 0.6
is working quite well for me.
On Dec 18, 2003, at 2:09 PM, mai
I think in this case the spammer has found a way to fool SA because SA
2.61 simply
won't process this message.
On Dec 15, 2003, at 9:56 AM, Robert Nicholson wrote:
I'm confused how this is suppose to work.
I've got this message that I think SA is ignoring.
I'm not gettin
I'm confused how this is suppose to work.
I've got this message that I think SA is ignoring.
I'm not getting any X-Spam headers at all over putting this message
thru SA.
>From [EMAIL PROTECTED] Mon Dec 15 00:51:53 2003
Received: (qmail 10603 invoked from network); 15 Dec 2003 06:51:41 -
Recei
I'm just curious in what order does eval_tests execute the tests?
Specifically are the offsite clearing houses only contacted whenever
my current score is less than required_hits?
ie. is there a way to conditionally do those offsite checks only if
they are necessary?
ie. my current score is less
Look at the body of the message and you should see a bunch of %'s in
the url.
You don't see them?
Does your MUA allow you to look at the raw text of the message?
On Dec 14, 2003, at 12:40 PM, Matt Kettler wrote:
At 01:24 PM 12/14/2003, Robert Nicholson wrote:
When you see encoded
When you see encoded urls like that you have to suspect that this is an
ebay phish right?
Does anybody know what the url is?
Received: (qmail 7380 invoked from network); 14 Dec 2003 00:45:49 -
Received: from 12-229-8-208.client.attbi.com (HELO earthling.net)
(12.229.8.208)
by grub.camro
Yes yesterday I enabled Net::DNS as well as adding DCC check and now
I'm pleased to say that spammer is no longer a concern.
I have a question a perl question and I'm not sure where in the various
Perl docs you are suppose to look for this but I want to know exactly
what this means.
package Ma
The way this interface is defined
$status = $f->learn ($mail, $id, $isspam, $forget)
suggests that both $isspam and $forget are significant however
it will only ever do one or the other not both.
---
This SF.net email is sponsored by: IBM Linu
So I'm executing
my $status = $spamtest->learn ($mail);
and the status in the debugger shows all the right things suggesting
it's been learnt
however the nham count in sa-learn --dump isn't changed.
---
This SF.net email is sponsored by: I
Is there anyway I can arrange for a message that DCC would catch to be
sent to me?
If you can do that just send it over I want to see if DCC is working
correctly.
---
This SF.net email is sponsored by: IBM Linux Tutorials.
Become an expert i
Hi,
Is this the correct place for SA to put installed when installed
locally and not system wide?
/home/robert/lib/perl5/site_perl/5.8.0/Mail
ie. does it belong in a 5.8.0 specific directory?
---
This SF.net email is sponsored by: IBM Linux
So I wanted uninstall 2.60 which is installed in a version specific
directory on my setup.
bash$ ls
SpamAssassin SpamAssassin.pm
bash$ pwd
/home/robert/lib/perl5/site_perl/5.8.0/Mail
However when I do this with the Makefile used to build it I get
bash$ make uninstall
no packlist file found:
nd an active checking account.
Get your advance today.
http://lc.orderfreebies.com/newlc/go/2746
Follow this link if you no longer wish to be contacted.
http://lc.orderfreebies.com/central/[EMAIL PROTECTED]
On Dec 13, 2003, at 12:00 AM, Bryan Hoover wrote:
Robert Nicholson wrote:
Anybody who's
Your double address was different if you compare it to the one below.
yours only has
one set of angle brackets.
Anyway I'm not suggesting that this is suitable criteria for
determining spam. It's just a reliable
signature for this particular spammer.
On Dec 12, 2003, at 10:09 PM, JRiley wrote:
Anybody who's installed A 2.61
Can you tell me what the score for the following email was
From: Advance in Pay <[EMAIL PROTECTED]>
Subject: Need cash in the bank ASAP?
Date: December 12, 2003 4:28:24 PM CST
To: Robert David Nicholson <[EMAIL PROTECTED]>
Reply-To: <[EMAIL PROTECTED]
ractive Homeowner News - December 2003
Date: December 6, 2003 10:23:27 AM CST
To: Robert Nicholson <[EMAIL PROTECTED]>
Reply-To: Loyalty Interactive
<[EMAIL PROTECTED]>
From: Hot Holiday Toy <[EMAIL PROTECTED]>
Subject:Get a Remote
In addition to SA I use a local baysien based filter and I'm wondering
if having X-Spam-Flag: YES as a header in all messages will cause this
local filter to not be able to correctly identify false positives set
by SA.
Is it possible to simply possible to invoke is_spam() and in the
process no
Any of this possible .. you just have to post process the headers of
the email
You would just look at the X-Spam-Status
eg.
X-Spam-Status: No, hits=0.0 required=0.6 tests=none autolearn=no
version=2.60
How you determine the score will depend on you are invoking SA.
On Tuesday, October 21, 20
My previous setup involved a combination of procmail, spamassassin call
from filtmail which is a perlscript that runs
from my .procmailrc.
I am reimplementing the lot in perl using SpamAssassin and Mail::Audit
with qmail.
so far everything works quite nicely.
However, in the past I used my own
Mail::SpamAssassin::NoAudit implements accept
Never mind I found this in it's accept implementation
# we don't support maildir or qmail here yet. use the real Mail::Audit
# for those.
On Saturday, October 4, 2003, at 12:06 AM, Rich Puhek wrote:
Robert Nicholson wrote:
Currently
folks here don't seem to like much) can
write out maildir's.
Q. As SA can be used as a Mail::Audit plugin via NoAudit can it also
write in maildir
format or only mailboxes?
On Friday, October 3, 2003, at 11:43 PM, Rich Puhek wrote:
Robert Nicholson wrote:
I'm calling SA from
er 3, 2003, at 10:41 PM, Rich Puhek wrote:
Robert Nicholson wrote:
Any Maildir afficionados here?
My ISP has just moved from mailboxes to Maildir with qmail.
Anybody know if qmail will support box folder types concurrently?
Likewise courier-imap is the imap server I need to communicate
with.
No, s
Any Maildir afficionados here?
My ISP has just moved from mailboxes to Maildir with qmail.
Anybody know if qmail will support box folder types concurrently?
Likewise courier-imap is the imap server I need to communicate
with.
---
This sf.net
Has anybody here used both Spamassassin and MailTools?
specifically has anybody tried to remove attachments from the Message
objects that SA provides?
But if I'm using Spamassassin's Mail::Audit plugin methods or
Mail:Audit's directly how can I then
specify the newer message that I built withou
I currently call perlscripts from Procmail and I invoke Spamassassin
within those perlscripts however I notice
that the developers of SA no longer trust Mail::Audit, at least
according to some of the comments in spamassassin
so what exactly are the concerns about using Mail::Audit instead of
pro
So really the crux of this task is to correctly reforward the email as
a *.txt attachment and
correctly identify said attachment before passing it to sa-learn.
The first part being the toughest. as it's the kind of thing you want
to automate or at least
have done for you via a keystroke. So thi
Just an idea to integrate sa-learn into the Mail (MUA) workflow. Since
I already invoke spamassassin from within perl I assume it's possible
to sa-learn a message via perl as well.
I suppose it should be possible to set up a service invoked via
procmail such that
you can email any spam that sho
So, just exactly where is the documentation that defines the criteria
for what gets auto learnt?
---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
___
Sp
Hi, have any improvements been made to sort out spammers who are
forging failure notices?
From: Mail Delivery Subsystem <[EMAIL PROTECTED]>
Date: Mon Sep 22, 2003 3:05:39 PM Asia/Bangkok
To: <[EMAIL PROTECTED]>
Subject: Returned mail: User unknown
Return-Path: <>
Delivered-To: [EMAIL PROTECTED]
Forgive me but do the # of hits actually have to be more than
required_hits?
I was under the impression that if I set required_hits to 6 then
whenever hits was greater than or equal to required_hits then that mail
would be considered spam.
From: "Mei Kwan NG" <[EMAIL PROTECTED]>
Date: Tue Sep
What was always_add_report set to?
I noticed in 2.55 even if the report was included in HAM the wording
still suggested the message was spam.
What was X-Spam-Status set to for this message that you claim was
interpreted as spam?
On Tuesday, September 23, 2003, at 06:58 PM, CHRISTOP TATRO wrot
It seems i didn't appreciate that autolearn in that line relates to how
that particular message was autolearnt.
So things are working as I expect them to.
---
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgee
RE:
X-Spam-Status:
Yes, hits=10.7 required=0.6 tests=DATE_IN_PAST_03_06,
DATE_SPAMWARE_Y2K,FORGED_MUA_OUTLOOK,FORGED_OUTLOOK_HTML,HTML_60_70,
HTML_FONTCOLOR_BLUE,HTML_FONTCOLOR_GREEN,HTML_FONTCOLOR_RED,
HTML_MESSAGE,MIME_HTML_ONLY,MIME_HTML_ONLY_MULTI,MISSING_MIMEOLE,
NORMAL_HTTP_TO_IP autolea
Never mind I just learnt about always_add_report
On Monday, September 22, 2003, at 08:19 PM, Robert Nicholson wrote:
In addition to the tests that were performed I'd like to see the
effect on the overall score each test had.
so instead of
X-Spam-Status: No, hits=-0.6 required=0.6
In addition to the tests that were performed I'd like to see the effect
on the overall score each test had.
so instead of
X-Spam-Status: No, hits=-0.6 required=0.6
tests=IN_REP_TO,NO_REAL_NAME,REFERENCES,X_AUTH_WARNING,X_LOOP
version=2.55
I'd like to see (-1.0) etc after each test. so that I
This problem I'm seeing is because folks are using my domain when
forging emails.
On Monday, September 22, 2003, at 06:49 PM, Ralf Guenthner wrote:
- Original Message -
From: "Robert Nicholson" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, Septemb
Hi, have any improvements been made to sort out spammers who are
forging failure notices?
From: Mail Delivery Subsystem <[EMAIL PROTECTED]>
Date: Mon Sep 22, 2003 3:05:39 PM Asia/Bangkok
To: <[EMAIL PROTECTED]>
Subject: Returned mail: User unknown
Return-Path: <>
Delivered-To: [EMAIL PROTECTED]
ok folks Spamassassin isn't doing a very good job with this guys spam
From: "Mark" <[EMAIL PROTECTED]>
Date: Tue Sep 16, 2003 11:16:00 AM Asia/Bangkok
To: [EMAIL PROTECTED]
Subject: I heard from them :)
Return-Path: <[EMAIL PROTECTED]>
Delivered-To: [EMAIL PROTECTED]
Delivered-To: [EMAIL PROTECT
I'm getting a lot of messages come back with -xx.x hits lately.
X-Spam-Status: No, hits=-22.2 required=1.0
tests=BAD_CREDIT,BAYES_20,CLICK_BELOW,FORGED_MUA_OUTLOOK,
IN_REP_TO,REFERENCES,USER_AGENT_MOZILLA_UA,USER_AGENT_PINE
autolearn=ham version=2.50
This was obvious spam that was sent me.
Ar
What is this counter suppose to reflect?
I use the sa-learn with the mbox option with a mailbox with 733 messages
this counter increases by one.
---
This SF.net email is sponsored by:Crypto Challenge is now open!
Get cracking and register here
55 matches
Mail list logo
