Re: SpamAssassin not hitting well on obvious spam

2007-10-16 Thread Jeff Chan
Quoting Chris 'Xenon' Hanson [EMAIL PROTECTED]: [...] X-Spam-Status: Yes, hits=4.4 required=4.0 X-Spam-Level: X-Spam-Report: SA TESTS 0.1 FORGED_RCVD_HELO Received: contains a forged HELO 0.1 HTML_40_50 BODY: Message is 40% to 50% HTML 0.0 HTML_MESSAGE

Re: SpamAssassin not hitting well on obvious spam

2007-10-16 Thread Theo Van Dinter
On Mon, Oct 15, 2007 at 11:53:09PM -0600, Chris 'Xenon' Hanson wrote: And yet, sometimes the spam that makes it through is startlingly obvious. Lots of expletives about male anatomy and the like, in plaintext mails. I turned on the X-Spam-Report header to see how things were going. A typical

Re: SpamAssassin not hitting well on obvious spam

2007-10-16 Thread Chris 'Xenon' Hanson
Jeff Chan wrote: Turn on SURBL tests. ecamn.com is blacklisted on SURBL. Ok. According to http://wiki.apache.org/spamassassin/SURBL http://www.surbl.org/faq.html#nettest SA 3.x have SURBL by default and it should be enabled if I'm not starting spamd with the -L/--local option. My

way to change the header

2007-10-16 Thread cpayne
Is there a way I can take this X-Spam-Status: No, score=(1.1), required=1.5, tests=BAYES_50,Magi_Body_Chuck, NO_RECEIVED,TO_CC_NONE, autolearn=no, bayes score = 0.5000, version=3.1.8 date scan = Mon, 15 Oct 2007 15:38:39 -0400 and make it look like this... X-Spam-Status: No,

Re: SpamAssassin not hitting well on obvious spam

2007-10-16 Thread Justin Mason
Henrik Krohns writes: http://taint.org/2007/08/15/004348a.html Ah, my auto-generated ruleset! Yes, please try it out -- it works very well indeed ;) (If anyone gets any FPs from it, I'd appreciate if you could package them up as an mbox, zip it, and mail it to me to avoid them in future. But

Re: way to change the header

2007-10-16 Thread Banyan He
I remember it follows RFC specification. So, no blank lines in the headers. cpayne wrote: Is there a way I can take this X-Spam-Status: No, score=(1.1), required=1.5, tests=BAYES_50,Magi_Body_Chuck, NO_RECEIVED,TO_CC_NONE, autolearn=no, bayes score = 0.5000, version=3.1.8 date scan

Re: SpamAssassin not hitting well on obvious spam

2007-10-16 Thread Loren Wilton
I believe SA uses Bayes out of the box, but what I don't get is how will Bayes know it's spam (to train on, versus ham) You tell it. Bayes won't kick in on a new installation until you have manually fed it AT LEAST 200 each hams and spams. You do this by deciding yourself if a message is

Re: Headers not being updated

2007-10-16 Thread MartyG
Thank you both for your posts I'll go off and find out about these issues. I'll report back if and when I find a solution. Martin Matt Kettler-3 wrote: MartyG wrote: I have recently moved to a new VPS, everything has been setup for me and is working well except Spamassassin. (I've

DNSWL question

2007-10-16 Thread Mark Wendt (Contractor)
I've started seeing some spam come through that gets labeled with RCVD_IN_DNSWL_MED RBL: Sender listed at http://www.dnswl.org/;, which imparts a negative score if the relay is listed in their db. Here at the Lab, we have an email gateway at the front, which is the single point of entry for

Re: DNSWL question

2007-10-16 Thread Jeff Chan
Quoting Mark Wendt (Contractor) [EMAIL PROTECTED]: I've started seeing some spam come through that gets labeled with RCVD_IN_DNSWL_MED RBL: Sender listed at http://www.dnswl.org/;, which imparts a negative score if the relay is listed in their db. Here at the Lab, we have an email gateway at

Re: DNSWL question

2007-10-16 Thread Jonathan Armitage
Mark Wendt (Contractor) wrote: I've started seeing some spam come through that gets labeled with RCVD_IN_DNSWL_MED RBL: Sender listed at http://www.dnswl.org/; Can someone point me in the right direction as to how and where I can turn off this rule if it can be turned off? In local.cf:

Re: DNSWL question

2007-10-16 Thread Justin Mason
Mark Wendt (Contractor) writes: I've started seeing some spam come through that gets labeled with RCVD_IN_DNSWL_MED RBL: Sender listed at http://www.dnswl.org/;, which imparts a negative score if the relay is listed in their db. Here at the Lab, we have an email gateway at the front,

AWL: what would happen if...

2007-10-16 Thread Mike Jackson
I work for a large managed hosting company. One of my fellow techs saw a customer put the following in their /etc/mail/local.cf: score AWL -100 He asked me (I have a reputation as the local SA expert) what would happen, and I couldn't figure it out. I figure it would either override the

Re: Mailman cooler [Was Re: unsubscribed]

2007-10-16 Thread SM
At 15:52 15-10-2007, Mark Martinec wrote: Also, not to forget that mailman in its current version invalidates and removes DKIM signatures, while this mailing lists stays faithful and keeps messages intact and retains original signatures. (there is supposedly some mailman patch floating around to

RE: uribl.com implementing ACLs

2007-10-16 Thread Chris Santerre
Since the last DDOS it would have been nice if the big guys ran local mirrors instead of making the problem worse. No donations and hammering away at the server I wonder why small RBLs drop off the planet. I salute every one who has donated time, machines, banwidth, and love to URIBL. The

Re: uribl.com implementing ACLs

2007-10-16 Thread John Rudd
IMO, one of the best and _easiest_ things any site can do to show love to any blacklist service is: run a local mirror. Even better is to run a publicly accessible mirror ... but a local mirror lessens your impact on the service you're consuming. Ask them when and often you can pull the

Re: AWL: what would happen if...

2007-10-16 Thread Theo Van Dinter
On Tue, Oct 16, 2007 at 10:44:45AM -0500, Mike Jackson wrote: happen, and I couldn't figure it out. I figure it would either override the dynamic score and always apply -100 if there's a match in the AWL database, or that the forced score would be ignored. So, which is it, or an answer C

refefine/extend a existing rule in 20_ratware.cf possible ??

2007-10-16 Thread Paul Griffith
Hi, I want to add a patch to 20_ratware.cf so I can extend FORGED_MUA_OUTLOOK to handle the new Outlook Message-ID format. Can I just redefine the supporting meta rule __FORGED_OE and let SA take care of the rest ? see patch here:

Re: Question about total effective of spamassassin

2007-10-16 Thread Alex Woick
I am also running an old version (3.1.7 on Ubuntu 7.04). Between SA and Thunderbird's own spam features, I am detecting something between 75% and 80% of spam. How much better is 3.2.x? On my small system (5 users) spam detection is above 99% accuracy for my own mail account. Less than 1

Sa-compile error

2007-10-16 Thread Luis HernĂ¡n Otegui
Hi, everybody, sa-compile was running allright in my systems, and the saturday it began to spit out this output (from sa-compile -D): cd /tmp/.spamassassin28680clJUyOtmp cd Mail-SpamAssassin-CompiledRegexps-body_0 Wide character in print at /usr/local/bin/sa-compile line 379, $fh line 4428. re2c

Re: Sa-compile error

2007-10-16 Thread Justin Mason
=?ISO-8859-1?Q?Luis_Hern=E1n_Otegui?= writes: Hi, everybody, sa-compile was running allright in my systems, and the saturday it began to spit out this output (from sa-compile -D): there are a number of bugs in sa-compile that are fixed in SVN trunk-- please apply the patches from

RE: uribl.com implementing ACLs

2007-10-16 Thread Joseph Brennan
No donations IT departments managed by folks with corporate backgrounds don't even have a procedure for sending off checks in arbitrary amounts solely because somebody thinks it would be a nice thing to do. Just say that large sites have to pay for rsync and put a price on it. That turns it

Re: refefine/extend a existing rule in 20_ratware.cf possible ??

2007-10-16 Thread Joseph Brennan
--On Tuesday, October 16, 2007 12:59 -0400 Paul Griffith [EMAIL PROTECTED] wrote: Hi, I want to add a patch to 20_ratware.cf so I can extend FORGED_MUA_OUTLOOK to handle the new Outlook Message-ID format. Can I just redefine the supporting meta rule __FORGED_OE and let SA take care of the

Re: unsubscribed

2007-10-16 Thread Clay Davis
Bob, I agree and have for a long time. I am always a little taken aback when an unsubscriber get hammered with sarcasm on this list... Plus, I have always assembled first and read the directions later... especially on Christmas Eve, when the pressure is on! It's human nature... but, then, so

Re: refefine/extend a existing rule in 20_ratware.cf possible ??

2007-10-16 Thread Paul Griffith
On Tue, 16 Oct 2007 14:57:00 -0400, Joseph Brennan [EMAIL PROTECTED] wrote: --On Tuesday, October 16, 2007 12:59 -0400 Paul Griffith [EMAIL PROTECTED] wrote: Hi, I want to add a patch to 20_ratware.cf so I can extend FORGED_MUA_OUTLOOK to handle the new Outlook Message-ID format. Can

RE: unsubscribed

2007-10-16 Thread Steve Ingraham
-Original Message- From: Clay Davis [mailto:[EMAIL PROTECTED] Sent: Tuesday, October 16, 2007 2:16 PM To: Bob Proulx; users@spamassassin.apache.org Subject: Re: unsubscribed Bob, I agree and have for a long time. I am always a little taken aback when an unsubscriber get

Re: refefine/extend a existing rule in 20_ratware.cf possible ??

2007-10-16 Thread Daryl C. W. O'Shea
Paul Griffith wrote: Hi, I want to add a patch to 20_ratware.cf so I can extend FORGED_MUA_OUTLOOK to handle the new Outlook Message-ID format. Can I just redefine the supporting meta rule __FORGED_OE and let SA take care of the rest ? Redefining the rule in your site config (often located

RE: unsubscribed

2007-10-16 Thread Clay Davis
Steve, I hope you didn't misunderstand me... I AGREE with you! Clay Steve Ingraham [EMAIL PROTECTED] 10/16/2007 4:10 PM -Original Message- From: Clay Davis [mailto:[EMAIL PROTECTED] Sent: Tuesday, October 16, 2007 2:16 PM To: Bob Proulx; users@spamassassin.apache.org Subject:

RE: unsubscribed

2007-10-16 Thread Steve Ingraham
-Original Message- From: Clay Davis [mailto:[EMAIL PROTECTED] Sent: Tuesday, October 16, 2007 3:33 PM To: users@spamassassin.apache.org Subject: RE: unsubscribed Steve, I hope you didn't misunderstand me... I AGREE with you! Clay No! My apologies for the misunderstanding.

Re: Question about total effective of spamassassin

2007-10-16 Thread cpayne
Matt Kettler wrote: Michael Scheidell wrote: 3.18 is unsupported. please update to latest versions. Well, it's as supported or unsupported as any other version of SpamAssassin is. No version of SpamAssassin is supported by the SA team beyond the advice given on this list. (sure,

Re: Question about total effective of spamassassin

2007-10-16 Thread Theo Van Dinter
On Tue, Oct 16, 2007 at 04:52:41PM -0400, cpayne wrote: but I have found out that sa-update places the lastest greatest updates in /var/lib/spamassasin and suse by default places the test files into /usr/share/spamassassin I think that is the problem. So know I have to figure out how to kill

Re: Question about total effective of spamassassin

2007-10-16 Thread Justin Mason
cpayne writes: Matt Kettler wrote: Thanks, guys the problem is that for SuSE 10.0 3.1.8 is the max, and the last time I update a 10.0 to the lastest everything broke. And the major thing for me that perl is still at 5.8.7, don't you have to be 5.8.8 or higher for the lastest stuff. nope.

RCVD_IN_DNSWL_LOW

2007-10-16 Thread Dan Mahoney, System Admin
dnswl.org is either full of it, or not well maintained. I've gotten at least 20 spams which I see are listed in dnswl.org as low trust (which still merits -1.0). Could we maybe please add a feature to spamassassin -r (or some other hook to the generic whitelisting code) which reports this to

40comcast.net detected as URI

2007-10-16 Thread Jim Hermann - UUN Hostmaster
The UUism Networks MailScanner believes that the attachment to this message sent to you From: [EMAIL PROTECTED] Subject: 40comcast.net detected as URI may be Unsolicited Commercial Email (spam). Unless you are sure that this message is incorrectly thought to be spam, please delete this

Re: unsubscribed

2007-10-16 Thread Bob Proulx
Steve Ingraham wrote: I just wanted to add my agreement to your statements and to ask that some posters try to treat all of us asking these supposedly stupid questions to understand that we really do struggle with understanding how all of these systems function. I see a lot of silly questions

Re: Question about total effective of spamassassin

2007-10-16 Thread cpayne
Justin Mason wrote: cpayne writes: Matt Kettler wrote: Thanks, guys the problem is that for SuSE 10.0 3.1.8 is the max, and the last time I update a 10.0 to the lastest everything broke. And the major thing for me that perl is still at 5.8.7, don't you have to be 5.8.8 or higher for the

Re: test my auto-generated ruleset

2007-10-16 Thread Larry Nedry
On 8/13/07 at 4:01 PM +0100 Justin Mason wrote: I've been working on a new way to auto-generate body rules recently... Are these rules restricted to Spamassassin 3.2 or newer? The following is what I get when I dig 8.1.3.sought.rules.yerp.org. Notice the NXDOMAIN. Thanks for the great work!

Re: Question about total effective of spamassassin

2007-10-16 Thread Matt Kettler
cpayne wrote: Cool I just found a source rpm and I am building 3.2.3.10 so hopefully this will help my issue thanks guys. By the way, I notices no one answer the part about the script. You mean this one: -- If questions, anyone know of script that works with

Re: DNSWL question

2007-10-16 Thread Matt Kettler
Justin Mason wrote: Mark Wendt (Contractor) writes: I've started seeing some spam come through that gets labeled with RCVD_IN_DNSWL_MED RBL: Sender listed at http://www.dnswl.org/;, which imparts a negative score if the relay is listed in their db. Here at the Lab, we have an email

Re: way to change the header

2007-10-16 Thread Matt Kettler
cpayne wrote: Is there a way I can take this X-Spam-Status: No, score=(1.1), required=1.5, tests=BAYES_50,Magi_Body_Chuck, NO_RECEIVED,TO_CC_NONE, autolearn=no, bayes score = 0.5000, version=3.1.8 date scan = Mon, 15 Oct 2007 15:38:39 -0400 Have you already customized your