The only time I've had problems with spamd dying (it was core dumping) was
because I was trying to run with the -m flag to keep things under control
in the event that something went wrong. After removing the -m switch, it
hasn't died since. Also, even when I do take down spamd, non of the e-mail
ge
> -Original Message-
> From: [EMAIL PROTECTED]
[mailto:spamassassin-talk-
> [EMAIL PROTECTED]] On Behalf Of Theo Van Dinter
> Sent: Monday, November 11, 2002 12:20 PM
> To: Apollo @ Carmel (lists account)
> Cc: [EMAIL PROTECTED]
>
> On Mon, Nov 11, 2002 at 01:49:17PM -0600, Apollo @ Carmel
Although there has been no talk of a solution, there continue to be people
who experience the problem of spamd dying.
The experience of someone else on this list pointed me toward the
'supervise' solution, but in the instances that spamd does die, it appears
that the mail is coming in fast enoug
I think I have found a perldoc problem:
perldoc Mail::SpamAssassin says:
$f->set_persistent_address_list_factory ($factoryobj)
Set the persistent address list factory, used to cre-
ate objects for the automatic whitelist algorithm's
persistent-storage back-end. S
Yes, SA does skip a test if the score is zero. In fact, this is exactly how
the standard ruleset skips the "for pay" RBLs.
At 06:41 PM 11/12/2002 -0600, Aaron Paetznick wrote:
Does SpamAssassin actually bother to run a check if it's score is set to
zero? I'm wondering if I can disable one or m
Except for the RCVD_IN_BONDEDSENDER rule, but then again that's secretly
run by the CIA as a front for their spaming operations in South America.
At 04:16 PM 11/12/2002 -0600, Jon Gabrielson wrote:
As far as I can tell, nonlocal tests only raise
the score, never lower it.
On Tue, Nov 12, 2002 at 04:59:51PM -0800, Matthew Cline wrote:
> RBL tests, and tests which involve the running of Perl code (eval tests)
> will be skipped if their score is set to zero. However, simply regular
> expression tests are always run.
Unless something's changed without me noticing (w
On Tuesday 12 November 2002 04:41 pm, Aaron Paetznick wrote:
> Does SpamAssassin actually bother to run a check if it's score is set
> to zero? I'm wondering if I can disable one or more individual RBL
> checks if I set their score to zero. Anybody?
>
> Thanks in advance!
RBL tests, and tests wh
Does SpamAssassin actually bother to run a check if it's score is set to
zero? I'm wondering if I can disable one or more individual RBL checks
if I set their score to zero. Anybody?
Thanks in advance!
--Aaron
---
This sf.net email is
I was looking for Bayesian filters and found this in CPAN.
Not sure how well it works or if it is event going to play nice
with Mail::Audit objects.
Does anyone have any experience or suggestions with the use of
Bayesian filters? I've seen a few comments here and there, but
nothing that shows
Tom Allison wrote:
> Klaus Heinz wrote:
> >A mistake (corrected in the most current CVS sources) disabled 'header'
> >rules specified in the users' configuration file. 'body' rules should
> >work, though. See also
> >
>
> This doesn't apply to Mail::SpamAssassin?
As the mistake happened in PerMs
sandra wrote:
HI,
I have run sa-learn-spam on my spam folders , but
when I forward a mail message again to me that
had already been learned by spamassassin, it scores
very low points to it (bayes_20, bayes_30) , but never
considers the message spam.
What is it? Do I have to define rules
Hello,
Been reading through the list for about a day now and I have a few
question. I am looking to setup SA as a DMZ to filter and then deliver to
our mail servers.
My question is dealing with user preferences. I know that you can setup
the user prefs in a SQL database, how about an LDAP backe
As far as I can tell, nonlocal tests only raise
the score, never lower it. If this is the case, it
might be nice to be able to skip nonlocal tests
if the score is already above the threshold.
Jon.
---
This sf.net email is sponsored by:
To lea
On Tue, Nov 12, 2002 at 09:52:53PM +, Martin Radford wrote:
> I was surprised that there are no rules to pick these up. I've had a
> few of these in the past.
>
> body T_INTL_DRIVERS_LICENSE /international driver'?s license/i
> describe T_INTL_DRIVERS_LICENSE International dr
At Tue Nov 12 17:51:12 2002, Jon Gabrielson wrote:
> Attached are 2 more false negatives.
...
> INTERNATIONAL DRIVER'S LICENSE
>
> Need a new driver's license?
I was surprised that there are no rules to pick these up. I've had a
few of these in the past.
I've submitted the following rule to
A bit more clear of a hint since not all of us use the same font types (ie:
on my MUA Theo mostly underlined SUB_FREE_OFFER):
score USER_IN_WHITELIST -100
So it would appear the from: address was manually whitelisted, or was one
of the few defaults that ships with SA.
Note that yo
Good help is hard to find.
I just want to thank Theo and the rest of the list for always offering
great help. This seems to be one of the better lists out there. Plus, I
love reading the Simpson's taglines at the bottom of Theo's posts :)
--
At 11:35 AM 11/12/2002, you wrote:
any idea why this happened. Look at the score:
X-Spam-Status: No, hits=-91.6 required=6.0
tests=BIG_FONT,CLICK_BELOW,CTYPE_JUST_HTML,DATE_IN_PAST_06_12,
EARN_PER_WEEK,EXPECT_TO_EARN,GREAT_OFFER,
HTML_COMMENT_UNIQUE_ID,HTML_FONT_COLOR_RED,
HTML_FONT_FACE_CAPS,INV
Justin Mason said:
> To examine how SpamAssassin is making its diagnosis, run with -D --
> this will print out debugging lines for each of the top tokens it
> uses to compute the overall result.
hmm, I should go into this in more detail: you want to keep an eye out
for lines like:
debug: bayes
On Tue, Nov 12, 2002 at 11:23:25AM -0800, Jason Winchell wrote:
> Local.cf contains:
> rewrite_subject 1
> report_header 0
>
> debug: is spam? score=2.2 required=5
> tests=DATE_MISSING,SPAM_PHRASE_00_01,TO_MALFORMED
>
> and these headers are inserted
> X-Spam-Status: No, hits=2.2 required=5.0
>
On Tue, Nov 12, 2002 at 02:35:04PM -0500, Phynias_CO wrote:
> any idea why this happened. Look at the score:
not without the whole message.
> X-Spam-Status: No, hits=-91.6 required=6.0
> SPAM_PHRASE_08_13,SUB_FREE_OFFER,USER_IN_WHITELIST
hint: ^
--
R
Having a bit of trouble with Spam Assassin. Here's the system specs.
Solaris 8
Perl 5.6.1
SpamAssassin 2.43
Local.cf contains:
rewrite_subject 1
report_header 0
The raw message being submitted is:
To: <[EMAIL PROTECTED]>
From: XXX <[EMAIL PROTECTED]>
Subject: Subject
X-Broadcast-ID: 7
X-Sent-To
any idea why this happened. Look at the score:
X-Spam-Status: No, hits=-91.6 required=6.0
tests=BIG_FONT,CLICK_BELOW,CTYPE_JUST_HTML,DATE_IN_PAST_06_12,
EARN_PER_WEEK,EXPECT_TO_EARN,GREAT_OFFER,
HTML_COMMENT_UNIQUE_ID,HTML_FONT_COLOR_RED,
HTML_FONT_FACE_CAPS,INVALID_DATE,LOW_PRICE,MONTH_TRIAL,
NO_
On Tue, Nov 12, 2002 at 01:05:43PM -0600, Patrick Bores wrote:
> header MY_SUBJECT_TEST Subject=~/subject1/
> header MY_SUBJECT_TEST Subject=~/subject2/
> header MY_SUBJECT_TEST Subject=~/subject3/
>
> and then have "score MY_SUBJECT_TEST -100" in a global file. This allows
>
> The way that spama
You need the Net::DNS perl Module for the MX check. And you need to
install Razor2 and DCC for those other tests.
You can d/l them at http://razor.sf.net and
http://rhyolite.com/anti-spam/dcc.
-Paul
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]] On Behalf Of Jon
Ga
Hello,
I'm trying to deploy SA site-wide and I'm writing a web configuration
tool.
What I would like to do is this:
Have the web configuration tool put the following into the user_prefs
file, like so:
header MY_SUBJECT_TEST Subject=~/subject1/
header MY_SUBJECT_TEST Subject=~/subject2/
header
sandra said:
>I have run sa-learn-spam on my spam folders , but
> when I forward a mail message again to me that
> had already been learned by spamassassin, it scores
> very low points to it (bayes_20, bayes_30) , but never
> considers the message spam.
have you run sa-learn-nonspam as well?
ok, i removed the -L option, but the following 3
rules still don't show up. Is there something special
I need to do to make these rules show up?
3.1 -- No MX records for the From: domain
3.9 -- Listed in Razor2, see http://razor.sf.net/
2.7 -- Listed in DCC, see http://rhyolite.com/anti-spam/dc
On Tue, Nov 12, 2002 at 01:48:47PM +0200, Vasco Macaringue wrote:
> Hi all,
> I'd like to get help and sugestions in installing and testing
> spamassassin . what milter do you sugest to use with sendmail
> 8.12.6 , knowing that my operating system is freeBSD 4.6.2
> RELEASE . From readme files of m
Attached are 2 more false negatives.
Jon.
p.s. is there a better place to report these?
>From [EMAIL PROTECTED] Mon Nov 11 22:20:37 2002
Return-Path: <[EMAIL PROTECTED]>
Delivered-To: [EMAIL PROTECTED]
Received: from zhuhai.mbfax.com (unknown [218.13.250.81])
by www.directfreight.com
HI,
I have run sa-learn-spam on my spam folders , but
when I forward a mail message again to me that
had already been learned by spamassassin, it scores
very low points to it (bayes_20, bayes_30) , but never
considers the message spam.
What is it? Do I have to define rules for every spam
> > I'm having trouble understanding how I should use the
> > methods compile_now() and init().
> >
> > Right now I use something like this.
> >
> > $spamtest->compile_now(1);
My way of thinking is that compile_now() should be used if you're going
to be checking multiple messages (>2)
On Tue, Nov 12, 2002 at 07:58:09AM -0800, Russ Gilman-Hunt wrote:
> I recall when I was setting up analog (http://analog.cx) it would expand
> files that ended in .gz. You could look at the setup scripts for analog and
> modify the spamstats.pl file to behave in a similar fashion. As an added
Lars Michael Jogback said:
> I'm having trouble understanding how I should use the
> methods compile_now() and init().
>
> Right now I use something like this.
>
> my $mail = Mail::SpamAssassin::NoMailAudit->new();
>
> my $spamtest = Mail::SpamAssassin->new();
>
>
On Tue, Nov 12, 2002 at 08:13:45AM -0500, Tom Allison wrote:
> OK, how do you report to RBL? Do you?
> I know Razor is through Mail::Audit::Razor...
It depends on the RBL. All of the ones I use require a mail in some
format specifying IPs. MAPS wants a copy of the offending message,
ordb and os
If I am not mistaken, you should only need to have a script that tails the
current maillog. Whatever data is found goes into the RRD and RRDtool will
handle aging out the historical data for you. You shouldn't need to parse
the old mail log files.
http://www.zcore.net/cacti/graph.php?graphid=100&r
I recall when I was setting up analog (http://analog.cx) it would expand
files that ended in .gz. You could look at the setup scripts for analog and
modify the spamstats.pl file to behave in a similar fashion. As an added
bonus you could post your modifications to the list. :^)
-Russ
On Tues
On Tue, Nov 12, 2002 at 09:54:35AM -0500, hou, ming wrote:
> Hi,
>
> I am running Spamassassin 2.43 with Perl 5.80 on Sun UltraSparc 500 MHz
> machine.
>
> I have tested with 500 concurrent e-mails sent to Sendmail through
> Spam-milter
> v1.2 to spamc, spamd, and Spamassassin(these processes we
Hi,
I am running Spamassassin 2.43 with Perl 5.80 on Sun UltraSparc 500 MHz
machine.
I have tested with 500 concurrent e-mails sent to Sendmail through
Spam-milter
v1.2 to spamc, spamd, and Spamassassin(these processes were running at the
same machine). I found that number of e-mails not able t
Theo Van Dinter wrote:
On Mon, Nov 11, 2002 at 01:49:17PM -0600, Apollo @ Carmel (lists account) wrote:
How do I make SA to send messages with certain score or origin to
[EMAIL PROTECTED]
Also, what do I have to enter in conf file (for whole system) to have
all e-mail originating from any host
Theo Van Dinter wrote:
On Sun, Nov 10, 2002 at 11:43:50AM -0700, Bob Proulx wrote:
Hi, I see that spamassassin -d will remove markup from a single message,
but is there a good way to remove spamassassin markup from an entire
mbox?
formail -s spamassassin -d < mailbox-tagged > mailbox-untagged
Klaus Heinz wrote:
Patrick Bores wrote:
Piping a message to spamassassin applies the rule correctly.
Piping the same message to spamc ignores the rule.
Any ideas as to why this is happening?
A mistake (corrected in the most current CVS sources) disabled 'header'
rules specified in the us
Hello,
I'm having trouble understanding how I should use the
methods compile_now() and init().
Right now I use something like this.
my $mail = Mail::SpamAssassin::NoMailAudit->new();
my $spamtest = Mail::SpamAssassin->new();
$spamtest->compile_now(1);
my
Hi all,
I'd like to get help and sugestions in installing and testing spamassassin .
what milter do you sugest to use with sendmail 8.12.6 , knowing that my
operating system is freeBSD 4.6.2 RELEASE .
>From readme files of mimedefang I knew that I can't use this milter or I'm
wrong
Please help me
No, I don't think that's such a good idea, I don't want a mailog that ends
up to large
We want the logs to be parsed even when they have been rotated so we can
retain the history otherwise, whats the point of the monthly or yearly
graph.
spamstats.pl needs to me modified so it can read /var/log
46 matches
Mail list logo
