At 08:51 PM 6/13/03 -0700, you wrote:
Anyone have a nice rule that will catch the attached? It got negative
scores with fake In-Reply-To, Approved-By, and X-Authentication-Warning
lines. Also, note the faked PGP signature with random words following
it. Sure is a lot of trouble to go through just t
-Original Message-
From: Benjamin A. Shelton [mailto:[EMAIL PROTECTED]
Sent: June 12, 2003 4:08 PM
To: [EMAIL PROTECTED]
Subject: Re: [SAtalk] spamassassin-qmail-vpopmail-qmailscanner - but no
spam scanning...
> What can I do to ensure that spamassassin is being
> used by qmail-scanner -
Anyone have a nice rule that will catch the attached? It got negative
scores with fake In-Reply-To, Approved-By, and X-Authentication-Warning
lines. Also, note the faked PGP signature with random words following
it. Sure is a lot of trouble to go through just to get a piece of spam
to my inbox.
--
> > The bayes database is global or for each user.
>
> It could be global or for each user. Better results when is applied for
> each
> user.
>
... ok.. and... global AND each user (both). It is posible??
Thanks
---
This SF.NET email
At 08:31 PM 6/13/03 +0200, Kai Schaetzl wrote:
You don't need to subscribe to send mail to this list? Isn't that
nonsense?
Read the archives.. this list is an open list, and it always has been..
This gets discussed at least once every 3 months, usually whenever a spam
message gets posted and some
Tony Earnshaw said:
> Tom Meunier wrote:
> > I'm kind of confused here. The way I see it (which could very well be a mi
> > sunderstanding, mind you) is that the reason it autolearns spam over 15 point
> > s by default is to make darned sure that it doesn't learn a false positive.
> > Then one
Tony Earnshaw wrote:
Simon Byrnand wrote:
Well, if you think other tests are not being run on base-64 encoded
messages, then you may have found a bug, because all the normal tests,
including bayes are supposed to run on the decoded text. (AFAIK)
People who (normally) are so dense that they refus
On Fri, 2003-06-13 at 18:00, Tony Earnshaw wrote:
> Jonathan Nichols wrote:
>
> > Is there a way to have a "whitelist_from" file instead of individual
> > entries in local.cf?
>
> /etc/mail/spamassassin/whitelist_from.cf
>
> In as much as any .cf file in that directory will be read in whenever
Simon Byrnand wrote:
Well, if you think other tests are not being run on base-64 encoded
messages, then you may have found a bug, because all the normal tests,
including bayes are supposed to run on the decoded text. (AFAIK)
People who (normally) are so dense that they refuse to give software and
Jonathan Nichols wrote:
Is there a way to have a "whitelist_from" file instead of individual
entries in local.cf?
/etc/mail/spamassassin/whitelist_from.cf
In as much as any .cf file in that directory will be read in whenever
whatever it is (spamd, amavisd-new etc) starts up. Given read rights,
Tom Meunier wrote:
I'm kind of confused here. The way I see it (which could very well be a
misunderstanding, mind you) is that the reason it autolearns spam over 15 points by
default is to make darned sure that it doesn't learn a false positive. Then one would
augment its learning by feeding
Hi,
Works globally without any hacks, but per user is a little more
difficult to implement.
I had promised a few month's ago to work on something to allow this but
time seems to have a way of vanishing before I can make any really
headway on it.
Regards,
Rick
[EMAIL PROTECTED] wrote:
This
This has probably been spoken about to death, but I can't google any
references to it...
Is there *any* way to use Bayes filtering when also using SQL preferences?
---
This SF.NET email is sponsored by: eBay
Great deals on office technology --
This came in with the SA version below, shall I report this via Bugzilla?
X-Mailer: The Bat! (v1.63 Beta/7)
X-Spam-Checker-Version: SpamAssassin 2.60-cvs (1.190-2003-06-01-exp)
X-Spam-Level: *
X-Spam-Status: Yes, hits=5.3 required=5.0 tests=FORGED_MUA_THEBAT
FORGED_MUA_THEBAT_BOUN autolearn=n
> I've heard (but don't know how true it is), that in some cases
> spammers send spam advertising anti-spam software in a bid to
> discredit the software and spread misinformation about it (and its
> authors).
Hey, it wouldn't surprise me! I guess that tells ya what they think about
Norton's soft
> I'm having great luck with spamassassin... a few things slip thru of
> course, and I wanted to post in case it gives developers a good idea.
>
> one thing I noticed is that for messages that are enc-64 , it seems that
> other tests are NOT run on the decoded message. I'd suggest decodeding
> and
>> This was a hoot...!
>
> This is one thing I'll never understand. Why advertise anti-spam software
> by spamming? I mean, if it really did work, you wouldn't be able to
> advertise anymore!
Actually it makes perfect sense in a sick twisted kind of way - once
people had bought their product, (a
Martin Radford said:
> At Fri Jun 13 17:14:50 2003, Benjamin A. Shelton wrote:
> >
> > > This was a hoot...!
> >
> > This is one thing I'll never understand. Why advertise anti-spam software
> > by spamming? I mean, if it really did work, you wouldn't be able to
> > advertise anymore!
>
> I'
At Fri Jun 13 17:14:50 2003, Benjamin A. Shelton wrote:
>
> > This was a hoot...!
>
> This is one thing I'll never understand. Why advertise anti-spam software
> by spamming? I mean, if it really did work, you wouldn't be able to
> advertise anymore!
I've heard (but don't know how true it is),
Sorry, I use both products together and subscribe to both lists.
FYI, I use MimeDefang as a milter within Sendmail-Switch to call
SpamAssassin.
Soon after I sent this I realized the mistake but it was too late for
recall.
You might try the mimedefang list -- this one's for SpamAssassin.
http://lists.roaringpenguin.com/mailman/listinfo/mimedefang
Charles Mount wrote:
I have exe in my $bad_exts list in the mimedefang-filter.
We have a person who can send *.exe files from his msn account and they do
not get blocked.
Stuart Gall wrote on Fri, 13 Jun 2003 16:34:49 +0300:
> I thought the OP had not use ok_languages at all ( i.e. he had the
> default which is ALL) and someone suggested using ok_languages it en
> to help with false positives.
>
I overlooked that, then, indeed, this option wouldn't help with fals
Hi everyone!
Is there a way to have a "whitelist_from" file instead of individual
entries in local.cf?
Thanks,
-Jonathan
---
This SF.NET email is sponsored by: eBay
Great deals on office technology -- on eBay now! Click here:
http://adfarm.me
Chris Santerre wrote on Fri, 13 Jun 2003 09:22:34 -0400:
> I use a rule called MY_EVIL for URLs that contain frequently used links by
> spammers. They don't send from this address but they use it for images and
> unsubscribe links and such.
>
I think that *is* what Dan talked about. Instead of ad
Matt Kettler wrote on Fri, 13 Jun 2003 11:37:47 -0400:
> I don't think you're subscribed to this mailing list,
>
You don't need to subscribe to send mail to this list? Isn't that
nonsense?
Kai
--
Kai Schätzl, Berlin, Germany
Get your web at Conactive Internet Services: http://www.conactive.
I have exe in my $bad_exts list in the mimedefang-filter.
We have a person who can send *.exe files from his msn account and they do
not get blocked.
When I test from a hotmail or netscape account the *.exe attachments are
being removed and the warning message is being added.
Any ideas?
Thanks
C
I'm kind of confused here. The way I see it (which could very well be a
misunderstanding, mind you) is that the reason it autolearns spam over 15 points by
default is to make darned sure that it doesn't learn a false positive. Then one would
augment its learning by feeding missed spams through
John,
Upon further investigation, it appears the author of qmail-scanner has this
to say about vpopmail:
[snip]
Q-S doesn't work with Vpopmail Vpopmail basically strips out environment
variables set within the tcpserver SMTP rules file - specifically the
QMAILQUEUE environment variable. As it is
> This was a hoot...!
This is one thing I'll never understand. Why advertise anti-spam software
by spamming? I mean, if it really did work, you wouldn't be able to
advertise anymore!
*sigh* Thankfully, SpamAssassin has caught all of the antispam ads I've
been getting.
I've said it before and
> I think I have qmail set up to use tcpserver
> my tcp.smtp file looks like this:
That looks right...
> Perhaps I don't have tcpserver running correctly or
> something like that? Actually it is definitely running.
If the rules were compiled with tcprules, you should be good to go (on the
other
Thanks for the reply Benjamin,
I think I have qmail set up to use tcpserver my tcp.smtp file looks like this:
# No Qmail-Scanner at all for mail from 127.0.0.1
127.:allow,RELAYCLIENT="",RBLSMTPD="",QMAILQUEUE="/var/qmail/bin/qmail-queue"
# Use Qmail-Scanner with SpamAssassin on any mail from the
If you are getting spam messages that have been tagged by spamassassin, and
would rather get your spam messages un-tagged, contact your ISP and ask
them to disable the product for your account.
I strongly suspect this is what your are experiencing, but you weren't
specific about what emails you
>It is completely automatic, you may want to change the threshholds, spam is ok but
>you dont get much ham by default.
>An easier way to seed the bayes is to whitelist some of your clients.
sounds good ... and where is all information stored ?
in "bayes_path" ? when i don't use the sa-learn
comm
On Fri, 2003-06-13 at 15:41, Murph wrote:
> If I turn on the Bayesian filter and set it to auto learn will this cause
> problems like false negatives/positives? Basically I would like it to start
> learning while I collect a couple 1000 SPAMs and HAMs to teach it. Just
> wondering if I should leave
If I turn on the Bayesian filter and set it to auto learn will this cause
problems like false negatives/positives? Basically I would like it to start
learning while I collect a couple 1000 SPAMs and HAMs to teach it. Just
wondering if I should leave it off for the time being if it will cause
troubl
If I turn on the Bayesian filter and set it to auto learn will this cause
problems like false negatives/positives? Basically I would like it to start
learning while I collect a couple 1000 SPAMs and HAMs to teach it. Just
wondering if I should leave it off for the time being if it will cause
troubl
please no more emails
---
This SF.NET email is sponsored by: eBay
Great deals on office technology -- on eBay now! Click here:
http://adfarm.mediaplex.com/ad/ck/711-11697-6916-5
___
Spamas
Trying to build SA on a RH 7.1 box and I get:
C compiler cannot create executables
I know some rpms are missing that are required to build executables but for
the life of me can't remember which. I thought it was:
libstdc++-devel-2.96-81.i386.rpm
but it is already on the box.
Would anybody ha
Hi,
Thanks for the tip, I had added
whitelist_to [EMAIL PROTECTED]
The user told me that his subject lines are not
getting altered but he still receives the X-Spam-**
headers in the emails which he does'nt like. Will the
allow_spam_to option get rid of the headers ?? if not
is there a way not t
On Fri, Jun 13, 2003 at 06:35:12AM -0700, Lydia Leong wrote:
> How do I fix this? I don't know exactly how many messages, spam, or ham,
> there have been, but if I could put in a reasonable estimate, that would
> be fine. Alternatively, forcing Bayes to turn back on would do, though
> possibly less
>Dear List,
>
>I'm new to Spamassassin and it's great. But I've got a
>little understanding problem, where the doc's and FAQ
>doesn't help. I want to implement Spamassassin with
>Bayes on a Postfix machine with amavisd-new.
>Since it's only a Gateway feeding the bayes db
>with spam and ham
- Original Message -
From: "Kai Schaetzl" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Friday, June 13, 2003 3:31 PM
Subject: Re: [SAtalk] Non English Email
> Stuart Gall wrote on Thu, 12 Jun 2003 23:10:27 +0300:
>
> > ok_languages it en
> > Would not affect italian or english emai
Who learned everything he knows about SA from the members who contribute to
this list..
:-)
At 07:40 AM Friday, 6/13/2003, Dan O'Brien wrote -=>
Special thanks to Ed Kasky for pointing me in the right direction.
Ed Kasky
Los Angeles, CA
. . . . . . . .
The washing machine in my head is on spin
Adam T. Bowen wrote on Thu, 12 Jun 2003 21:12:42 +0100 (BST):
> An obvious exclusion I know but I just wanted to make sure as there is no
> way of telling SpamAssassin the local-host-names (sendmail style). Could
> a --please-ignore-all-the-lovely-headers-except-the-ones-I-tell-you switch
> po
wrote on Fri, 13 Jun 2003 10:52:39 +0200:
> So I only want to use "auto_learn". auto_learn is explained
> in the man-page of sa-learn, but when I only want to use
> auto_learn do I have to use this command or will all be
> done automatic ? And when yes, I an initial bayes db
> build required
I made the mistake of interrupting an expiry run with a ctrl-C. This
had the effect of zeroing out my bayes_msgcount file, and check_bayes_db
now shows:
0.0000 740 non-token data: nspam
0.000050 non-token data: nham
0.0000503150
Stuart Gall wrote on Fri, 13 Jun 2003 01:41:13 +0300:
> Is there a way to incorporate custom eval tests such that they will survive upgrades.
>
You can put all custom stuff in /etc/local/spamassassin.
Kai
--
Kai Schätzl, Berlin, Germany
Get your web at Conactive Internet Services: http://www
Hello,
I have got a number of e-mail viruses coming through my sendmail, what
anti-virus do you recommend?
rgds,
Frank
---
This SF.NET email is sponsored by: eBay
Great deals on office technology -- on eBay now! Click here:
http://adfarm.
> -Original Message-
> From: Dan Baker [mailto:[EMAIL PROTECTED]
> Sent: Friday, June 13, 2003 8:30 AM
> To: [EMAIL PROTECTED]
> Subject: [SAtalk] a few suggestions
>
>
> I'm having great luck with spamassassin... a few things slip thru of
> course, and I wanted to post in case it give
> Hi everyone,
>
> I am a newbie to spamassassin but managed to install
> it and am successfully running it right now. At my
> work place we have multiple servers on which we would
> like to install spamassassin to filter out SPAM. We do
> not have a centralised mail server where filtering is
>
I'm having great luck with spamassassin... a few things slip thru of
course, and I wanted to post in case it gives developers a good idea.
one thing I noticed is that for messages that are enc-64 , it seems that
other tests are NOT run on the decoded message. I'd suggest decodeding
and then run
Stuart Gall wrote on Thu, 12 Jun 2003 23:10:27 +0300:
> ok_languages it en
> Would not affect italian or english email scoring, if you recieved
> japanese email it would trigger the undesired_language rule.
>
That's just what I am saying. We were talking about Italian. "it" was
missing from ok_l
I am trying to run the Bayes Learner on a directory of spam that ive saved
from outlook express, there are 2000 spam messages in the folder, 1850-1900
are tagged by spam assassin which i believe doesnt cause a problem. outlook
express saves files as ".eml" when it saves them
yet when i ran the lea
Hi everyone,
I am a newbie to spamassassin but managed to install
it and am successfully running it right now. At my
work place we have multiple servers on which we would
like to install spamassassin to filter out SPAM. We do
not have a centralised mail server where filtering is
done. Right now i
By George, I think I've got it!
I did a bit of Googling trying to find out what the non-local SA checks
were. They include RBL -- which I'm not (intentionally) doing, but are
enabled by default. I've added "skip_rbl_checks 1" to my local.cf, and my
checks are taking <10secs.
Special thanks t
Title: auto_learn without "sa-learn"-feeding ?
Dear List,
I'm new to Spamassassin and it's great. But I've got a
little understanding problem, where the doc's and FAQ
doesn't help. I want to implement Spamassassin with
Bayes on a Postfix machine with amavisd-new.
Since it's only a Gat
Hi
> >3. Should I let it up to the users to create the rules? Or should
> the >system-admins make them for the users?
>
> Well this is really up to you, as you're asking us to define your
> policy for you but in general, give your users freedom of
> control, don't lock them into one thing or
57 matches
Mail list logo
