I've done extensive testing and now understand what's going on...but I don't
know how to fix it.
When MailScanner is running, all mail sent from any of the user accounts (on
my virtual domains), will get stuck on the server. In fact, it doesn't even
go into a queue directory anywhere. Each time a
On Mon, 14 Jul 2003, Luiz Lima wrote:
> How can I tell if bayes is actually working on my setup? I have a
> amavisd-new + sa setup and, although my bayes database gets updated and is
> checked when I run SA on it own, I'm yet to see any bayes rules being
> applied to e-mails I personally get (and
On Tue, 15 Jul 2003, Simon Byrnand wrote:
> >I had several spams not too long ago in which all of them used the same
> >Message-ID. SA refused to learn from any of them except the first one.
> >What can be done to combat that? Of course, duplicate Message-ID's are
> >a violation of RFC's, but sp
Abigail Marshall <[EMAIL PROTECTED]> wrote:
>
> Try putting the recipe that is in your personal .procmailrc at the end
> of the main /etc/procmailrc - then the mail will be sent to the
> $SUSPECT file BEFORE going to your individual POP boxes.
This would work, but it implies that the $SUSPECT file
At 00:29 14/07/03 -0500, Fuzzy Fox wrote:
Simon Byrnand <[EMAIL PROTECTED]> wrote:
>
> * Are you learning a proportionate amount of ham as well ? [...]
> * Bayes doesn't "learn" particular messages, it learns the statistics
> of the words used in the messages. [...]
None of this addresses the or
Hi,
Osirusoft and bl.spamcop.net are back up here.
Regards.
Rick
Tony Earnshaw wrote:
Tim wrote:
P.s: your score was much to low, considering your 2.60. You need to
start using RBL and DCC checks at the least (avoiding anything to do
with OSIRUS, at the moment - DoS death, namely.)
Sorry,
Jul 14 00:00:42 lonestar spamc[73960]: connect() to spamd at 207.135.64.67
failed, retrying (1/3): Connection refused
Jul 14 00:00:43 lonestar spamc[73960]: connect() to spamd at 207.135.127.67
failed, retrying (2/3): Invalid argument
Jul 14 00:00:44 lonestar spamc[73960]: connect() to spamd at 2
Tim wrote:
P.s: your score was much to low, considering your 2.60. You need to
start using RBL and DCC checks at the least (avoiding anything to do
with OSIRUS, at the moment - DoS death, namely.)
Sorry, I am using a single machine on a 56K dialup connection, and network
checks can be very time-
Matt Kettler wrote:
I think what happened is a few of the more "clever" spammers got his
address as a result of them mining addresses out of dice.com.
Of course, the original email itself asks no questions and provides no
information aside from stating that he's getting more spam that doesn't
Kris Deugau wrote:
I have a long and growing list of custom rules for message elements
found in spams that slipped through SA. Among others, I have a set of
rules for URIs. The regexes for most look like:
/(domain1|domain2|domain3|domain4)\.(com|net|org)/i
However, in the false-negative message
Matt Kettler wrote:
> I tested and couldn't reproduce the bug on 2.54
I figured that might be the case. :(
> I was using following rules (one each of uri, body and rawbody)
[snip]
> uri SPAM_SITE_11/(domainsforpeople|pandabearperks)\.com/i
Just for kicks, I tweaked the message to use th
At 12:37 PM 7/14/2003 -0700, Matt Thoene wrote:
Huh? One has absolutely nothing to do with the other. What kind of spam
is getting through?
I think what happened is a few of the more "clever" spammers got his
address as a result of them mining addresses out of dice.com.
Of course, the original em
At 03:42 PM 7/14/2003 -0400, Kris Deugau wrote:
I've attached the message I'm playing with right now (as a zip because
SourceForge rejected it the first time); here's the rule
that's failing to trigger:
uri SPAM_SITE_11/(domainsforpeople|pandabearperks)\.com/i
describe SPAM_SITE_11 Body
Hi Tony,
On Mon, 14 Jul 2003, Tony Earnshaw wrote:
> P.s: your score was much to low, considering your 2.60. You need to
> start using RBL and DCC checks at the least (avoiding anything to do
> with OSIRUS, at the moment - DoS death, namely.)
Sorry, I am using a single machine on a 56K dialup
Matt Thoene wrote:
I've installed spamassassin and it was working great up
until I signed up on dice.com. Dice.com is basically
a job board where I posted my resume. Since then a lot
of spam has been getting through spamassassin.
Huh? One has absolutely nothing to do with the other. What kind of
Only issue I could think of with Dice.Com is them selling your name to
spammers or crawlers finding your email address on Dice.Com ... maybe you
weren't getting much spam before you signed up, but now you are getting so
much it's noticeable?
But I agree with Mr. Thoene on this one; they have absol
I seem to have lost 6 hours worth of mail because I screwed up my procmail
stuff. If you answered me I apologise. Can we try again?
--
-Time flies like the wind. Fruit flies like a banana. Stranger things have -
-happened but none stranger than this. Does your driver's license say Organ
-Donor?B
Just viewing my mail logs today because i started getting fetchmail
errors saying I had duplicate processes running which usually singals
that I've got a stuck email somewhere. In the process of checking the
mail logs I saw this pop up as SA was trying to process a message it
later identified as s
I've been adminning SA (currently 2.55) on a server at work for almost a
year now, and I've yet to have any serious trouble with it. (Longer on
a private server.)
However, I've just today run into some *very* odd behaviour.
I have a long and growing list of custom rules for message elements
foun
At 01:19 PM 7/12/2003 -0700, I Am Jesus @sent wrote:
I suggest something like the original comments followed by:
#score RCVD_IN_BL_SPAMCOP_NET 1.8
#score RCVD_IN_DUL 0.8
#score RCVD_IN_DUL_FH 0.8
#score RCVD_IN_RBL 1.8
#score RCVD_IN_RSS 1.8
Flipping back to your original post...
I *think* what yo
On Monday, July 14, 2003 @ 10:57:29 AM [-0700], Alaw Guo wrote:
> I've installed spamassassin and it was working great up
> until I signed up on dice.com. Dice.com is basically
> a job board where I posted my resume. Since then a lot
> of spam has been getting through spamassassin.
Huh? One has
At 11:37 AM 7/14/2003 -0600, Eric wrote:
I setup the local.cf in my /etc/mail/spamassassin/local.cf with the
line required_hits 1.00
for testing purposes... however after killing spamd and restarting the
spam hits say 5.0 is the required_hits cutoff
my score changes in local cf are worki
perl -MCPAN -e shell
o conf prerequisite_policy follow
install Mail::SpamAssasssin
t/basic_lintok
t/db_awl_path...ok
t/db_based_whitelist_ipsok
t/db_based_whitelistok
t/forged_rcvd...ok
t/gtube.ok
t/lang_pl_tests.
Hi,
Apparently GEE Whiz is a Novell Groupwise filtering widget, a commercial
product that uses SpamAssassin rules (somehow):
"New features in this release include: Corporate Disclaimers/Signatures
(HTML and text, top and/or bottom of email), Realtime Black Lists (RBL)
with SPAM values, extensive
On Mon, 2003-07-14 at 18:25, Chris Yuzik wrote:
> Hi everyone,
>
> I have SpamAssassin and MailScanner running on my RedHat 7.3 machine.
> The server hosts about a dozen domains. I run sendmail and procmail.
> Almost everything works great. Almost.
>
> Mail from the outside world works as expec
How can I tell if bayes is actually working on my setup? I have a
amavisd-new + sa setup and, although my bayes database gets updated and is
checked when I run SA on it own, I'm yet to see any bayes rules being
applied to e-mails I personally get (and I do get a lot of e-mails).
Is there any sign
Hello,
I've installed spamassassin and it was working great up
until I signed up on dice.com. Dice.com is basically
a job board where I posted my resume. Since then a lot
of spam has been getting through spamassassin.
---
This SF.Net email s
I setup the local.cf in my
/etc/mail/spamassassin/local.cf with the
line
required_hits 1.00
for testing purposes... however after killing spamd and
restarting the spam hits say 5.0 is the required_hits cutoff
my score changes in local cf are working fine?
So what am I missing.
I
I Am Jesus @sent wrote:
Ok, so that means the behaviour is intentional, so the bug is in FAQ
1.5. I will edit a couple FAQ items and email JM. That should solve
the problem.
[...]
Umm...I tried to make it obvious that knew that.
Might try changing your handle and learning English in the mean
Hi everyone,
I have SpamAssassin and MailScanner running on my
RedHat 7.3 machine. The server hosts about a dozen domains. I run sendmail and
procmail. Almost everything works great. Almost.
Mail from the outside world works as expected, gets
scanned and scored and so forth. I've got SA's
- Original Message -
From: "Martin Bene" <[EMAIL PROTECTED]>
To: "Mark" <[EMAIL PROTECTED]>; "Fuzzy Fox" <[EMAIL PROTECTED]>;
<[EMAIL PROTECTED]>
Sent: Monday, July 14, 2003 1:08 PM
Subject: Re: [SAtalk] Re: Why wil "sa-learn" not learn?
> > Hi Mark,
> >
> > {root} % /usr/local/sa/bin/sa-
Nichols, William wrote:
Has anyone made a rule for dealing with comments in the middle of text,
a new one I just started getting is below:
[...]
Pretty bad, if anyone has one can you reply with it here?
These are not even comments, simply obfuscation.
The rules / function have been there for a lo
"Lucas Albers" <[EMAIL PROTECTED]> wrote:
I made up a new meta rule.
It will raise the value for any mail that matches both razor and dcc.
Based on your email it appears the false positive rate for mail that
matches both razor and dcc is very very low.
Actually it was Razor and Pyzor that I checked
Tim wrote:
Looks like we have some competition.
At least our developers can write English. Maybe Spam Remedy could be
persuaded to take English lessons from the developers? Dan? Matt?
Justin? People? Pocket money?
Best,
Tony
P.s: your score was much to low, considering your 2.60. You need to
Theo Van Dinter wrote:
On Sat, Jul 12, 2003 at 01:19:34PM -0700, I Am Jesus @sent wrote:
The default scores are 0 for several RBLs. Is this policy or is it just
that no one who participates in mass-check uses these RBLs, a bug of sorts?
The ones you specified aren't free. So they get di
Running version 2.55 for both spamc and spamd,
redhat 7.3
I can only reproduce the long timeout
values when the global procmail can’t connect to spamd so there are multiple
spamc processes running trying to connect, but if I run from the command like spamc
–d it times out fine.
T
Looks like we have some competition.
;-)
Tim
-- Forwarded message --
Received: from localhost [127.0.0.1] by localhost.localdomain
with SpamAssassin (2.60-cvs 1.193-2003-06-13-exp);
Mon, 14 Jul 2003 10:44:32 -0400
From: [EMAIL PROTECTED]
To: Kc8hr <[EMAIL PROTECTED
Has anyone made a
rule for dealing with comments in the middle of text, a new one I just started
getting is below:
DO YOU
WANNA BE
SHOCKED?!..
size=5>Y0UNGEST RUSSlAN
TEENs
GET NAKED and
FUCKED ONLY FOR
YOU!!!
- HARD TEEN P0RN0 --
FlRST TlME SE}
At 07:38 PM 7/13/2003 -0400, Mike Leone wrote:
> And who will pay for the software?
No one - you use one of the free anti-virus scanners.
agreed, clamav is free and work fine for this...
However a better question would be "who will pay for the hardware
upgrades".. Seeing as the server that runs
Hello,
I'm installing Spam Assassin on Win2K workstation (Active Perl 5.6.1)
Trying to install Time::HiRes and get this error.
PPM> install Time::HiRes
Install package 'Time-HiRes?' (y/N): y
Installing package 'Time-HiRes'...
Error installing package 'Time-HiRes': Could not locate a PPD file
On 07/13/03 21:58 -0400, John P Verel wrote:
>
> Also, evolution-1.2.2-5.
>
> John
Examination of my maillog shows these two entries:
Jul 13 09:50:11 Verdi spamd[21524]: identified spam (10.2/5.0) for
mary:501 in 0.2 seconds, 3865 bytes.
Jul 13 13:18:13 Verdi spamd[1433]: identified spam (17.3/
Steven,
Try putting the recipe that is in your personal .procmailrc
at the end of the main /etc/procmailrc - then the mail will
be sent to the $SUSPECT file BEFORE going to your individual
POP boxes.
-Abigail
Sunday, July 13, 2003, 3:41:50 PM, you wrote:
SWO> I have an account on my server for
Hi Mark,
>{root} % /usr/local/sa/bin/sa-learn --spam --mbox -D /tmp/spam.txt
>...
>debug: Initialising learner
>Learned from 0 messages.
At a guess it's trying to learn from a file called "-D" now. try moving the
-D parameter.
Bye, Martin
---
- Original Message -
From: "Fuzzy Fox" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, July 14, 2003 7:43 AM
Subject: [SAtalk] Re: Why wil "sa-learn" not learn?
> However, the original poster already stated that he used "sa-learn
> --forget" to try to remove any traces of the me
- Original Message -
From: "Fuzzy Fox" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, July 14, 2003 7:43 AM
Subject: [SAtalk] Re: Why wil "sa-learn" not learn?
> Simon Byrnand <[EMAIL PROTECTED]> wrote:
>
> > * Are you learning a proportionate amount of ham as well ? [...]
> > *
Never
thought about that... only tested when the host was up. However, I only get a
10-12 second timeout now when testing against a downed IP. I'm then running
the spamc service in a postfix configuration. I'm using v2.55 on a RedHat 8 box.
This is what I get in the logs:
Jul 14
10:26:23
46 matches
Mail list logo
