Title: Message
Great,
that worked well.
Somehow I missed this option in the Config man page? I must have searched
for TRUSTED instead of trusted. thanks for pointing it out!
jae
-Original Message-From: Andrew W. Donoho
[mailto:[EMAIL PROTECTED] Sent: Friday, October 15, 20
At 02:22 PM 10/15/2004, Johannes Rumpf wrote:
i'm runing xfilter "spamassassin" in my procmail rules. Thats the second
time the server runs spamassassin. First with spamd with an other points
and bayes filter, second via xfilter. The mails are detected correctly as
spam, but there are no x-spam*
At 04:51 PM 10/15/2004, Matt wrote:
Normally spamassassin works great.. but this one e-mail seems to cause
it to take forever (600 seconds?) and then get passed through with a
score of (?/?) <-- I assume that is because something timedout and
passed it along?
What's going on?
Have you tried turning
Hi, I use spamassassin3.0 on Debian GNU/Linux, invoked from postfix in
this way:
master.cf:
smtp inet n - - - - smtpd -o content_filter=spamassassin
[cut]
spamassassinunix- n n - 5 pipe
user=nobody argv=/usr/bin/spamc -f -e /usr/sbin/sendmail -oi -f
${sende
At 03:31 PM 10/15/2004, Daniel Olmedilla wrote:
In fact he is right in the fact that the autowhitelist is kept in memory.
However, it keeps only the e-mail addres and two numbers for each row so
assuming for example an average of 50 bytes per e-mail address (quite long)
and two numbers, even if you
On Fri, Oct 15, 2004 at 04:51:29PM -0400, Matt wrote:
> Normally spamassassin works great.. but this one e-mail seems to cause
> it to take forever (600 seconds?) and then get passed through with a
> score of (?/?) <-- I assume that is because something timedout and
> passed it along?
> What's goin
Normally spamassassin works great.. but this one e-mail seems to cause
it to take forever (600 seconds?) and then get passed through with a
score of (?/?) <-- I assume that is because something timedout and
passed it along?
What's going on?
Return-Path: <[EMAIL PROTECTED]>
Delivered-To: [EMAIL P
At 03:33 PM 10/15/2004, Michael Barnes wrote:
> That said, I agree with Andrew.. By disabling the rule you're just
ignoring
> a part of a larger problem. Your trust path is broken, and needs fixing
> ASAP. This affects things other than just the ALL_TRUSTED rule.
I'm not sure what you mean that m
Thank you Mark for your checks.
In fact he is right in the fact that the autowhitelist is kept in memory.
However, it keeps only the e-mail addres and two numbers for each row so
assuming for example an average of 50 bytes per e-mail address (quite long)
and two numbers, even if you had 100.00
Michael Barnes wrote:
> >I've got a rule that lookes similar to:
> >
> >body HAS_XX/^(?:>\s{0,3}){0,4}XX$/
> >describe HAS_XXThere looks like a XX here.
> >score HAS_XX-1.0
Matt Kettler wrote:
> you need to make that a rawbody rule and it should work.
Not necessary, but usef
On Fri, Oct 15, 2004 at 03:25:10PM -0400, Matt Kettler wrote:
> At 01:05 PM 10/15/2004, Michael Barnes wrote:
> >I've found that the ALL_TRUSTED hit too many spams as hams. I havn't
> >looked at the rule to see what it is doing, but I put in my
> >/etc/mail/spamassassin/local.cf
> >
> >ALL_TRUSTED
At 01:05 PM 10/15/2004, Michael Barnes wrote:
I've found that the ALL_TRUSTED hit too many spams as hams. I havn't
looked at the rule to see what it is doing, but I put in my
/etc/mail/spamassassin/local.cf
ALL_TRUSTED 0
Are you sure you didn't do this instead:
score ALL_TRUSTED 0
The orig
On Oct 15, 2004, at 12:05, Michael Barnes wrote:
I have a spam which hits ALL_TRUSTED. I've attached the "spamassassin -D
I've found that the ALL_TRUSTED hit too many spams as hams. I havn't
looked at the rule to see what it is doing, but I put in my
/etc/mail/spamassassin/local.cf
ALL_TRUSTED
At 02:10 PM 10/15/2004, Michael Barnes wrote:
I've got a rule that lookes similar to:
body HAS_XX/^(?:>\s{0,3}){0,4}XX$/
describe HAS_XXThere looks like a XX here.
score HAS_XX-1.0
Which is supposed to match things like:
> > > XX
you need to make that a rawbody rule and it sho
At 01:46 PM 10/15/2004, Mario Sergio Candian wrote:
If i send a spam mail with the string in the message body without any
subject:
GTUBE-STANDARD-ANTI-UBE-TEST-EMAIL
of my account to the server with SA, will i receive an email marked SPAM?
Yes, you it should tag it. You can even include whatever su
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Ray writes:
> There are bugtrack entries for the ALL_TRUSTED problem that you are
> describing. On my own network we were seeing all spam hit with -3.3 on
> ALL_TRUSTED. We are using SA on Postfix as a "man in the middle" relay from
> our AV to our
There are bugtrack entries for the ALL_TRUSTED problem that you are
describing. On my own network we were seeing all spam hit with -3.3 on
ALL_TRUSTED. We are using SA on Postfix as a "man in the middle" relay from
our AV to our main mail server.
MAIL <--> Postfix+SA <--> AV <--> Inernet
The cu
Ernie Dunbar wrote:
I'm running Spamassassin v3.0.0 on FreeBSD 4.9-RELEASE, and I run spamd like
this:
/usr/local/bin/spamd -A 127.0.0.1 -L -x -u spamc -d -m 10
When I use `ps aux |grep spamd` I get this:
spamc 61970 0.5 3.9 21292 20032 ?? Is3:04PM 0:00.47
/usr/local/bin/spamd -A 12
Hi @all,
i'm runing xfilter "spamassassin" in my procmail rules. Thats the second
time the server runs spamassassin. First with spamd with an other points
and bayes filter, second via xfilter. The mails are detected correctly
as spam, but there are no x-spam* in the mail-header so i can't move t
I've got a rule that lookes similar to:
body HAS_XX/^(?:>\s{0,3}){0,4}XX$/
describe HAS_XXThere looks like a XX here.
score HAS_XX-1.0
Which is supposed to match things like:
> > > XX
or
XX
etc. where the XX is at the end of the line. In other words, a XX at
the end o
> yeah, me too. maybe we can get one or two standing beside the
> new logo ;)
Agreed.. When I heard about the new logo, I had high hopes for a new
ninja look.. My heart fell when I saw that the ninjas were gone...
Don't get me wrong, the new logo is really nice, but I just loved
those ninjas..
> I have a spam which hits ALL_TRUSTED. I've attached the "spamassassin -D
I've found that the ALL_TRUSTED hit too many spams as hams. I havn't
looked at the rule to see what it is doing, but I put in my
/etc/mail/spamassassin/local.cf
ALL_TRUSTED 0
to disable the test. Maybe someone who is mo
Potato Chip wrote:
Thank you everyone for your input and for directing me to the real
problem -- SPF. For now, I have had to score ALL_TRUSTED -0.01 but would
still like to get to the bottom of this SPF, TRUSTED issue.
Actually, I suspect that what you're seeing with SPF is a symptom, not
the caus
On Fri, 2004-10-15 at 06:55, Chris Santerre wrote:
> Is this not at the mercy of how users put your email into their address
> book? What if my wife adds me to her address book under "Sexy stud muffin"?
> Would that not fail on this rule?
That would fail on many, many levels.
--
Calvin Dunigan
N
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Chris Santerre writes:
> >-Original Message-
> >From: ChupaCabra [mailto:[EMAIL PROTECTED]
> >Sent: Thursday, October 14, 2004 11:56 AM
> >Cc: SpamAssassin Users
> >Subject: Re: OT: Ninja Blanket
> >
> >
> >
> >
> >Kenneth Porter wrote:
> >
>
At 11:19 AM 10/15/2004, Martin Hepworth wrote:
hmm great, perl malware :-)
More seriously, anyone checked the code for nasties...
Disclaimer: I'm no perl expert, so treat my analysis that of someone with
limited experience.
I inspected the code and saw nothing terribly suspicious in my limited
u
Thank you everyone for your input and for directing me to the real
problem -- SPF. For now, I have had to score ALL_TRUSTED -0.01 but would
still like to get to the bottom of this SPF, TRUSTED issue.
I have a spam which hits ALL_TRUSTED. I've attached the "spamassassin -D
< spam" output below. I'v
At 11:23 AM 10/15/2004, Vee Persaud wrote:
Hi,
I was wondering why the following email was not tagged with a higher
score, yet it was quarantined. The email and my qmail-scanner options are
as follows:
*** Qmail-Scanner Quarantine Envelope Details Begin ***
X-Qmail-Scanner-Mail-From: "[EMAIL PR
Thanks so much. Unfortunately, I don't see much change in my CPU usage by
spamd. I am
at a loss, as I've spent almost an entire day reading old mailing list threads
and the
wiki, but no one has seemed to post anything concrete as to why spamd would eat
so much
CPU (in my case, it spikes to as
Matt Kettler wrote:
> Fair enough, then you know and understand that you're using an
> unstable, in-development version.
Indeed. It is a deliberate choice I made, and I do *not* intend to
complain about code currently under development not being tested as
thoroughly as official releases. No worries
Humm. I didn't expect this :-).
I don't know what is "aka" but of course you can check my publications list at
http://www.l3s.de/~olmedilla/pub/publications.html
As you can see I have been working in e-learning and in information retrieval
(with ranking algorithms and personalized web search). N
Chris Frederick wrote:
> _HITS_ gets replaced with
> the score, and _AUTOLEARN_ doesn't get changed at all.
H...
There are two possibilities that come to my mind.
(1) _AUTOLEARN_ isn't recognized by your version for some reason.
(2) The replacement mechanism is only working on the first tag fo
Hi,
I was wondering why the following email was not tagged with a higher score, yet
it was quarantined. The email and my qmail-scanner options are as follows:
Received: from mx1.magmacom.com (206.191.0.217)
by 198.103.10.100 with SMTP; 14 Oct 2004 18:17:52 -
Received: from webmail2.magma
At 10:41 AM 10/15/2004, Hou, Ming wrote:
Argument "n.nn" isn't numeric in addition (+) at
/usr/local/lib/perl5/site_perl/5.8.4/Mail/SpamAssassin/Conf.pm line 238.
Any idea? How to fix it?
Looks like you have an invalid score statement in one of your config
files
either it's non-numeric, or y
hmm great, perl malware :-)
More seriously, anyone checked the code for nasties...
--
Martin Hepworth
Snr Systems Administrator
Solid State Logic
Tel: +44 (0)1865 842300
Daniel Olmedilla wrote:
Dear all,
I am a Ph.D. student that works in Hanover (Germany). I am currently studying
the distributio
Quoting Alex Broens <[EMAIL PROTECTED]>:
[Daniel's request]
> aka harvesting?
His request does seem a bit shady, but I'd google Dan's previous papers before
jumping to conclusions.
--
Sahil Tandon
Daniel Olmedilla wrote:
Dear all,
I am a Ph.D. student that works in Hanover (Germany). I am currently studying
the distribution of spam mails and e-mail addresses. For that I am gathering
some information from e-mail distributions of institutions and also
individuals. I developed a script that
Dear all,
I am a Ph.D. student that works in Hanover (Germany). I am currently studying
the distribution of spam mails and e-mail addresses. For that I am gathering
some information from e-mail distributions of institutions and also
individuals. I developed a script that gathers the information
On Fri, Oct 15, 2004 at 10:41:57AM -0400, Hou, Ming wrote:
> I got a problem when I ran SpamAssassin 3.0's sa-learn. The error
> message was like:
> Argument "n.nn" isn't numeric in addition (+) at
> /usr/local/lib/perl5/site_perl/5.8.4/Mail/SpamAssassin/Conf.pm line 238.
You have a score in one o
>-Original Message-
>From: Jim Maul [mailto:[EMAIL PROTECTED]
>Sent: Friday, October 15, 2004 10:09 AM
>To: Chris Santerre
>Cc: users@spamassassin.apache.org
>Subject: Re: [RD] incorrect real name (was Re: feeding frenzy for
>ws.surbl .org!!!)
>
>
>Chris Santerre wrote:
>
>
>
>>
>>
>> I
That's perhaps what I'm after. Most spam that we receive seems to
originate from the US. While we do have a couple of US based clients,
they could be whitelisted if necessary.
Thanks for the tip.
Ralph
> -Original Message-
> From: Matt Kettler [mailto:[EMAIL PROTECTED]
> As far as I can
Title: Argument "n.nn" isn't numeric in addition (+)??
Hi,
I got a problem when I ran SpamAssassin 3.0’s sa-learn. The error message was like:
Argument "n.nn" isn't numeric in addition (+) at /usr/local/lib/perl5/site_perl/5.8.4/Mail/SpamAssassin/Conf.pm line 238.
Any idea? How to fix
At 08:49 AM 10/15/2004 -0400, Robert Benites wrote:
I am having a problem with my installation of Spamassassin 3.0 on my
RedHat EL/WS system. I've googled the problem and found some reference
to a similar problem from users running RedHat 9. Trying the proposed
solution does _not_ fix the problem f
At 11:56 AM 10/15/2004 +0200, Ralph Seichter wrote:
I'm using Subversion to access the latest SpamAssassin code.
Fair enough, then you know and understand that you're using an unstable,
in-development version. A lot of the code is being freshly tested here and
may not work perfectly, though it ge
Matt Kettler wrote:
At 08:12 PM 10/15/2004, Chris Frederick wrote:
SA version is 2.63. And I tried setting report_safe to 0, 1, and 2
and nothing changed. I changed it in /etc/mail/spamassassin/local.cf
and my ~/.spamassassin/user_prefs, even restarting the daemon between
changes (not sure if
Chris Santerre wrote:
Is this not at the mercy of how users put your email into their address
book? What if my wife adds me to her address book under "Sexy stud muffin"?
Well then she'd probably be lying ;)
-Jim
>-Original Message-
>From: John Wilcock [mailto:[EMAIL PROTECTED]
>Sent: Thursday, October 14, 2004 5:19 AM
>To: users@spamassassin.apache.org
>Subject: [RD] incorrect real name (was Re: feeding frenzy for
>ws.surbl.org!!!)
>
>
>On Thu, 14 Oct 2004 08:51:56 +0200, John Wilcock wrote:
>> I
>-Original Message-
>From: ChupaCabra [mailto:[EMAIL PROTECTED]
>Sent: Thursday, October 14, 2004 11:56 AM
>Cc: SpamAssassin Users
>Subject: Re: OT: Ninja Blanket
>
>
>
>
>Kenneth Porter wrote:
>
>> Was browsing ThinkGeek just now (looking for the Swiss Army
>USB memory
>> fob) and noti
>-Original Message-
>From: Theo Van Dinter [mailto:[EMAIL PROTECTED]
>Sent: Thursday, October 14, 2004 11:45 AM
>To: users@spamassassin.apache.org
>Subject: Re: Default SURBL scores low?
>
>
>On Thu, Oct 14, 2004 at 08:27:02AM -0700, Potato Chip wrote:
>> -3.3 ALL_TRUSTED Most of these un
Use spamstats:
http://freshmeat.net/projects/spamstats/
and run against old and new maillogs.
--Chris
>-Original Message-
>From: Ronan [mailto:[EMAIL PROTECTED]
>Sent: Friday, October 15, 2004 9:27 AM
>To: spam
>Subject: comparison
>
>
>Is there any way to compare the effectiveness or o
Is there any way to compare the effectiveness or otherwise or 2.6x vs
/3.0.0??
I have just upgraded one of my 3 mailhubs to 3.0.0 and want to compare
it to the other 2 in terms of effectiveness...
how would i go about this or is there no way?
ronan
--
Regards
Ronan McGlue
==
Analyst
On Fri, 15 Oct 2004, Scott George wrote:
> I was wondering if anybody knows of a script that pulls SA information
> from the syslog, like pflogsumm does for Postifx?
>
> I could write my own, but I don't want to re-invent and existing wheel
> if there is one.
Check out spamstats.
--
Mike Burge
I am having a problem with my installation of Spamassassin 3.0 on my
RedHat EL/WS system. I've googled the problem and found some reference
to a similar problem from users running RedHat 9. Trying the proposed
solution does _not_ fix the problem for my installation.
Before asking a question, I wan
I was wondering if
anybody knows of a script that pulls SA information from the syslog, like
pflogsumm does for Postifx?
I could write my
own, but I don't want to re-invent and existing wheel if there is
one.
TIA
Scott George
Customer Support Engineer
Digital Controls
Corporation
Hi folks,
I'm using Subversion to access the latest SpamAssassin code. Lately,
"spamassassin --lint" returns this error:
config: warning: score set for non-existent rule DIGEST_MULTIPLE
Here ist what the rule files show:
$ grep DIGEST_MULTIPLE /usr/share/spamassassin/*
20_body_tests.cf:meta D
On Tuesday 12 Oct 2004 22:05, Joe Jenkins wrote:
> I am trying to make Spam Assassin 3.0 redirect all tagged spam emails to
> one file in /var/spool/SPAM instead of ever having it show up in the
> individual users email boxes. I've got spamassassin working now (thanks,
> theo and matt) and it see
On Sun, 2004-10-10 at 20:04, jdow wrote:
Someone used my address with the name Daniel Hofer. So about 10% of the
spam I get is instantly filterable with that name and variants.
Well, a new variant of 419 has hit my mailbox. Someone named Allan Hofer
died in Nigeria. And he left a big estate.
Bill,
I guess this is easy to say, but I would dig out an old
machine and have a go at getting Linux up and running and adding Postfix (or
some other MTA) plus Spamassassin.
I'm also a "windows server guy" but I found it pretty
straightforward with the info on this and other mailing lists
>
> At 07:24 AM 10/14/2004, BG Mahesh wrote:
>
>
>
> >How do I fix this problem?
>
>
> Remember when the "perl Makefile.pl" warned you that checking network rules
> could cause make test to fail, but you turned them on anyway?
>
Yes, I turned it on. On 3 machines I had installed it witho
>
> Hi,
>
> On Thu, 14 Oct 2004 16:54:04 +0530 "BG Mahesh" <[EMAIL PROTECTED]> wrote:
>
> > I am trying to install SA 3.0 on RedHat linux. During testing the error
> > I get is,
> >
> > % make test
> > [dnsbl test failures]
> >
> > How do I fix this problem?
>
> What does:
>
> perl -e 'use
Justin Mason writes:
> there is one alright -- bug 3826. could you check and see if that
> matches what you're seeing?
>
> http://bugzilla.spamassassin.org/show_bug.cgi?id=3826
That's it, exactly. The spamd children don't drop the required_score from
user_prefs and apply it to later messages.
I
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Loren Wilton writes:
> > Has anyone else experienced this problem? I will try to set up a separate
>
> Yep, you aren't the first. 2-3 other people have commented on this. I
> don't know if there is an official bug on this particular aspect or not;
> Has anyone else experienced this problem? I will try to set up a separate
Yep, you aren't the first. 2-3 other people have commented on this. I
don't know if there is an official bug on this particular aspect or not;
there are some related problems that do have open bugs. I'm also not sure
i
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
64MB should be plenty -- those files shouldn't stick around once
a scan has completed, and they should all be 250KB or less in
size.
In my experience tmpfs helps quite a lot, but in this case it's
a trivial amount of the overall total runtime.
- --j
> > I have been under the apparently false presumption that spamd prcessed
> > its messages in memory (perhaps this explains why each spamd process can
> > oft take up to 25% cpu?). I recently looked in /tmp and found lots of
> > left over spamassassin..xx.tmp files (the first four x's are
>
> Oh - also, sorry for the self-followup...
>
> normal ps output:
> root 8978 1 0 17:55 ?00:00:00 /usr/bin/perl
> -T -w /usr/local/bin/spamd -d -D
> root 8979 8978 0 17:55 ?00:00:00 spamd child
> root 8980 8978 0 17:55 ?00:00:00 spamd child
> root
Oh - also, sorry for the self-followup...
normal ps output:
root 8978 1 0 17:55 ?00:00:00 /usr/bin/perl -T -w
/usr/local/bin/spamd -d -D
root 8979 8978 0 17:55 ?00:00:00 spamd child
root 8980 8978 0 17:55 ?00:00:00 spamd child
root 8981 8978
At 08:12 PM 10/15/2004, Chris Frederick wrote:
SA version is 2.63. And I tried setting report_safe to 0, 1, and 2 and
nothing changed. I changed it in /etc/mail/spamassassin/local.cf and my
~/.spamassassin/user_prefs, even restarting the daemon between changes
(not sure if it was needed but wa
On Tue, Sep 28, 2004 at 11:56:52AM -0700, Justin Mason wrote:
> Will Yardley writes:
> > On Fri, Sep 24, 2004 at 10:22:01AM -0700, Justin Mason wrote:
> > > Will Yardley writes:
> > > > Has anyone else seen a problem w/ spamd dying sometimes (after
> > > > working for a while)? I have been seeing
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
email builder writes:
> Greetings,
>
> I have been under the apparently false presumption that spamd prcessed
> its messages in memory (perhaps this explains why each spamd process can
> oft take up to 25% cpu?). I recently looked in /tmp and found
Matt Kettler wrote:
At 06:44 PM 10/14/2004, Matt Kettler wrote:
>Is there any reason that the _AUTOLEARN_ tag isn't getting replaced
with "no", "ham", or "spam"? Or is there something with my
distribution's setup that's causing this?
If report_safe is not set to 0, SA won't do that, it will onl
> On Thu, Oct 14, 2004 at 08:37:05AM +0200, Erik Slooff wrote:
> > Only recently I've seen this error:
> > [08:31] [EMAIL PROTECTED]:/home/erik > sa-learn --sync
> --force-expire
> > Odd number of elements in hash assignment at
> >
> /usr/local/lib/perl5/site_perl/5.8.3/Mail/SpamAssassin/BayesSt
I'm running SpamAssassin 3.0.0 on a Red Hat Linux 9 system (2.4.27). The
mail server is running sendmail 8.12.11.
Users turn spamassassin on or off for their accounts by using procmail
rules to call spamc. There are many users doing this, and it has worked
perfectly until now. When I upgraded t
73 matches
Mail list logo
