awesome! looks like it removes addresses seen only once. it also seems to be
okay with SA 3.0.
Thanks much! (Kris, too:)
-Original Message-
From: snowjack [mailto:[EMAIL PROTECTED]
Sent: Friday, October 08, 2004 4:21 PM
To: users@spamassassin.apache.org
Subject: Re: AWL auto_expire?
Nate Schindler wrote:
Just a curiosity question for now - is auto-expiring the AWL a planned
feature?
My auto-whitelist is about 3x the size of bayes_toks. I imagine it'll
become problematic eventually, since it's only growing.
...or is there already some way to expire old entries from the AWL,
Good to know - I was not aware that it did the --sync after learning.
Ed
At 03:54 PM Friday, 10/8/2004, Nate Schindler wrote -=>
a --sync operation is performed when you sa-learn things. This commits
the journal to the database, and removes the file.
it's recreated, written to, committed, and rem
Hi,
I followed the instructions in the ReportingSpam and RazorSiteWide Wiki
pages to try to enable spam reporting on my server. I'm using RedHat 9,
Postfix, SpamAssassin 3.0 and Razor 2.61. I have SpamAssassin set-up to
run from /etc/procmail and it works fine, including the razor2 checks
and so
Title: AWL auto_expire?
Just a curiosity question for now - is auto-expiring the AWL a planned feature?
My auto-whitelist is about 3x the size of bayes_toks. I imagine it'll become problematic eventually, since it's only growing.
...or is there already some way to expire old entries from t
a --sync operation is performed when you sa-learn things. This commits the
journal to the database, and removes the file.
it's recreated, written to, committed, and removed automatically when needed.
sa-learn just forces this to happen whenever it's run.
it's by design.
-Original Message-
Hey all --
I've been considering possible improvements to how we figure out what
rules are effective.
Currently we use the S/O ratio and hit-rate of each individual rule, in
other words, if a rule hits a lot of spam, and little nonspam, we detect
that and consider it "good".
However, that doesn'
I am currently running SA 3.0.0 with a site wide bayes and spamd running as
user spamd.
Database is in /home/spamd
When I pipe false negatives through sa-learn, the bayes_journal file
disappears. Is this by design or is there something I need to change or fix?
Thanks...
Ed
. . . . . . . .
"It'
Thanks guys, problem was i commented out the old stuff:
#:0fw
#| /usr/bin/spamassasin
#:0
when i needed to leave the second :0 uncommentedd'oh!
thanks again
> From: Morris Jones <[EMAIL PROTECTED]>
> Date: Fri, 8 Oct 2004 13:19:01 -0700 (PDT)
> To: "[EMAIL PROTECTED]" <[EMAIL PROTECTED]>
>
[EMAIL PROTECTED] wrote:
Thank you, appeared to work great, logging to maillog and all, the problem
now becomes where it's putting spam.
previously my procmailrc looked like this:
:0fw
| /usr/bin/spamassassin
:0
* ^X-Spam-Status: Yes
$HOME/spam
which routed spam to the users spam folder
when i chan
On Fri, 8 Oct 2004, Michael Parker wrote:
> Of all the folks seeing memory issues, are you using ok_languages in
> your config somewhere? If not, please speak up as well.
Yes:
ok_languagesen
Mojo
--
Morris Jones <*>
Monrovia, CA
[EMAIL PROTECTED]
http://www.whiteoaks.com
Michael Parker wrote:
On Tue, Oct 05, 2004 at 12:25:45PM -0500, Michael Parker wrote:
On Tue, Oct 05, 2004 at 10:22:42AM -0700, Morris Jones wrote:
I watched a spamd child grow to 250MB yesterday on a single message. I
have a suspicion that the memory usage growth is happening on a whiteli
You have two separate things happening in two recipes:
The first recipe:
> :0fw
> | /usr/bin/spamassassin
Adds the X-SpamStatus: header to the email. The second recipe:
> :0
> * ^X-Spam-Status: Yes
> $HOME/spam
Delivers email with the specified header to $HOME/spam.
The problem I was address
On Tue, Oct 05, 2004 at 12:25:45PM -0500, Michael Parker wrote:
> On Tue, Oct 05, 2004 at 10:22:42AM -0700, Morris Jones wrote:
> >
> > I watched a spamd child grow to 250MB yesterday on a single message. I
> > have a suspicion that the memory usage growth is happening on a whitelist
> > or bayes
Thank you, appeared to work great, logging to maillog and all, the problem
now becomes where it's putting spam.
previously my procmailrc looked like this:
:0fw
| /usr/bin/spamassassin
:0
* ^X-Spam-Status: Yes
$HOME/spam
which routed spam to the users spam folder
when i change it to:
:0fw
* < 25
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
>
> Thank you Matt!
>
> You're info is great, but I'm sure you dont mind if I take your
> disclaimer seriously.
>
> Before I go on, anyone that uses spamd confirm this info...?
His suggestions sound right on to me. For the most part all you nee
His information is all correct.
Change your procmailrc lines so they're something like this:
# Pass through spamassassin
:0fw
* < 256000
| /usr/bin/spamc
Mojo
On Fri, 8 Oct 2004, [EMAIL PROTECTED] wrote:
> Thank you Matt!
>
> You're info is great, but I'm sure you dont mind if I take your dis
Sahil Tandon wrote:
>
snowjack wrote:
Is there any evidence that local.cf is getting read at all?
Good question. Where do I look for such evidence? Logs are hardly
revealing.
If you first log in, or su to be the user amavis is running as, then run
"spamassassin -D --lint" you should see somethi
[EMAIL PROTECTED] wrote:
Quoting [EMAIL PROTECTED]:
It seems that the marked spam is never getting to this point, to have these
rules applied to it. It is being marked and then being sent to the user,
marked as spam, so that part is working.
If the headers are there, SA is doing its job. Ask th
Thank you Matt!
You're info is great, but I'm sure you dont mind if I take your disclaimer
seriously.
Before I go on, anyone that uses spamd confirm this info...?
> From: Matt Kettler <[EMAIL PROTECTED]>
> Date: Fri, 08 Oct 2004 13:48:40 -0400
> To: "[EMAIL PROTECTED]" <[EMAIL PROTECTED]>,
> S
At 01:16 PM 10/8/2004, [EMAIL PROTECTED] wrote:
Yea, ive been asked this before. I inherited this setup, so Im trying to
work that out.
i can restart SA by using rc.d/init.d/spamassassin
it shows in ps aux as /usr/bin/spamd -d -c -a
and when mail comes in it shows as
/usr/bin/perl -T -w /usr/bin/s
Michael Parker wrote:
Hi All,
I'm compling some stats on Bayes and AWL in SQL for my upcoming talk
at ApacheCon[1]. If you're making use of either Bayes and/or AWL with
SQL based storage could you please respond (off list is fine) to this
email. I will only be using raw numbers, no names. In parti
Yea, ive been asked this before. I inherited this setup, so Im trying to
work that out.
i can restart SA by using rc.d/init.d/spamassassin
it shows in ps aux as /usr/bin/spamd -d -c -a
and when mail comes in it shows as
/usr/bin/perl -T -w /usr/bin/spamassassin
is that helpful?
> From: Ma
At 12:55 PM 10/8/2004, [EMAIL PROTECTED] wrote:
Im sure SA should be logging something to somewhere, probably maillog right?
Well I dont have anything in the maillog from SA. Am I missing something?
Anyone help me debug why it's not logging?
How are you calling SA?
If you're using spamc/spamd then
Hey all:
running SA 2.6 on Linux with Sendmail.
Im sure SA should be logging something to somewhere, probably maillog right?
Well I dont have anything in the maillog from SA. Am I missing something?
Anyone help me debug why it's not logging?
Hi All,
I'm compling some stats on Bayes and AWL in SQL for my upcoming talk
at ApacheCon[1]. If you're making use of either Bayes and/or AWL with
SQL based storage could you please respond (off list is fine) to this
email. I will only be using raw numbers, no names. In particular, I'm
interested
Hi,
On Fri, 8 Oct 2004 10:33:08 -0500 [EMAIL PROTECTED] wrote:
[...]
> It seems that the marked spam is never getting to this point, to have these
> rules applied to it. It is being marked and then being sent to the user,
> marked as spam, so that part is working. Is there another file somewhere
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Kai Schaetzl writes:
> The problem seems to exists on all of our Bayes databases and I think the
> cause is not "bad" data, but simply the way the SA expiry algorithm works.
> There are no negative atimes or atimes in the future. If the database
>
Quoting [EMAIL PROTECTED]:
> It seems that the marked spam is never getting to this point, to have these
> rules applied to it. It is being marked and then being sent to the user,
> marked as spam, so that part is working.
If the headers are there, SA is doing its job. Ask the procmail mailing
wrote on Fri, 8 Oct 2004 10:33:08 -0500:
> I cannot configure spamassassin to delete spam.
>
sa doesn't do this! Do you mean to send mail to /dev/null with procmail?
> procmailrc
>
why not ask the procmail list?
http://MailMan.RWTH-Aachen.DE/mailman/listinfo/procmail
Kai
--
Kai Schätzl,
The problem seems to exists on all of our Bayes databases and I think the
cause is not "bad" data, but simply the way the SA expiry algorithm works.
There are no negative atimes or atimes in the future. If the database
contains tokens from a wide time range it's not able to calculate a
reasonab
More of the same...
(spamassassin 2.64 on freeBSD 5.2.1)
I appreciate the help I have recieved so far, but it still doesn't work...
and yes, I have read the wiki, and I have googled the hell out of it...
I cannot configure spamassassin to delete spam. I have tried many, many
combinations of the
At 08:01 AM 10/8/2004, Sean Doherty wrote:
Spamassassins DCC configuration option "use_dcc" specifies
whether to use DCC or not. However, it appears that
Spamassassin will perform a dcc check if dccifd is available
(if the socket specified under dcc_dccifd_pathor exists) or
use_dcc is set to 1. The
Matt Kettler wrote:
> Theoretically, it should also match if the hostname changes, as long
> as the domain+TLD part is the same (ie: foo.blah.com)
>
> That said I've heard some mumblings the SA 3.0 implementation
> of domain stripping is a bit different than the
> Mail::SpamCopURI version, and the
At 08:43 AM 10/8/2004, Matt wrote:
I have a question on the new(ish) scanning that spamassassin does on
URI's.It seems to be working very well for us here, but I have a
question..
WIll it catch:
http://www.blah.com/?jj38942
as well as
http://www.blah.com/?34223
The URI blacklists only check the
On Thu, 7 Oct 2004, Jeff Tucker wrote:
> I captured an exact copy of one of the messages that was being scanned
> when this happened.
> [ ... ]
> Rescanning the same message by calling spamc didn't cause the
> problem. The scan completed in just a couple of seconds.
I did exactly the same expe
Hi,
On Fri, 08 Oct 2004 12:30:59 +0200 Cedric Foll <[EMAIL PROTECTED]> wrote:
> First I've had a look on my spam scores and i saw a strange behavior,
> BAYES_99 get a lower score (1.9) than BAYES_95 (2.0).
> I've had a look on http://spamassassin.apache.org/tests_3_0_x.html and
> this score seem
Quoting Keith Hackworth <[EMAIL PROTECTED]>:
> I have the same configuration and I had the same problem with 2.64. I did
> a spamassassin --lint and it showed me all sorts of problems in my
> local.cf. It was so bad, it just ignored anything beyond x lines in the
> file.
Very plausible - I'll t
Quoting Volker Kindermann <[EMAIL PROTECTED]>:
> if you are running postfix,amavisd-new and spamassassin chrooted be sure to
> not only change /etc/mail/spamassassin/local.cf but also
> /etc/mail/spamassassin/local.cf
I'm not running within a chroot; the cf is in /usr/local/etc/mail/spamassassin.
It only looks at the domain.
For example, it will catch:
www.blah.com/
anything.blah.com
anything.anything.more.subdomains.blah.com/anything
Keith
> Hi,
> I have a question on the new(ish) scanning that spamassassin does on
> URI's.It seems to be working very well for us here, but I have a
Hi,
I have a question on the new(ish) scanning that spamassassin does on
URI's.It seems to be working very well for us here, but I have a
question..
WIll it catch:
http://www.blah.com/?jj38942
as well as
http://www.blah.com/?34223
We are beginning to notice alot of e-mails (being marked ri
Hi all,
I have SA 3.0 rc2 running on my Slackware 9.0 box, and want to
upgrade to the official 3.0 release. I looked through the UPGRADE
file and didn't see any information on whether I need to do anything
with my Bayes databases during the process.
Thanks for any guidance.
--
Theodore (Ted)
Hi,
Spamassassins DCC configuration option "use_dcc" specifies
whether to use DCC or not. However, it appears that
Spamassassin will perform a dcc check if dccifd is available
(if the socket specified under dcc_dccifd_pathor exists) or
use_dcc is set to 1. The same logic is in both 2.64 and 3
I have the same configuration and I had the same problem with 2.64. I did
a spamassassin --lint and it showed me all sorts of problems in my
local.cf. It was so bad, it just ignored anything beyond x lines in the
file.
I'm not sure why, but my install of sa was VERY picky about spacing in the
fi
ADMIN_miki wrote:
>
> this machine is located inside DMZ
> any ideas why are these problems ?
> Thank you
> Miki
The obvious question I would ask is are you allowing connections outgoing ?
ie. Is your firewall open on the required ports?
Mr Michele Neylon
Blacknight Internet Solutions Ltd
Host
Hi,
I try "make test" in SA 3.0
and in part of dnsbl test I got these errors>
[EMAIL PROTECTED] Mail-SpamAssassin-3.0.0]# make test
PERL_DL_NONLAZY=1 /usr/bin/perl "-MExtUtils::Command::MM" "-e" "test_harness(0,
'blib/lib', 'blib/arch')" t/*.t
t/basic_lintok
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi,
First I've had a look on my spam scores and i saw a strange behavior,
BAYES_99 get a lower score (1.9) than BAYES_95 (2.0).
I've had a look on http://spamassassin.apache.org/tests_3_0_x.html and
this score seem normal when use of network tests.
But
Dear all!
I am using SA (2.63, SuSe Linux 9.0) in a German language environment to get the
score description etc. out in German (I have LANG=de_DE LANGUAGE=de_DE export
LANG LANGUAGE in my /etc/sysconfig/spamd)
Because of this, SA also seems to writes a German Date-Header into the taged
mails i
Eugene Morozov wrote:
> Loren Wilton wrote:
>>> I'm wondering, would it be useful to have a plugin that penalizes
>>> messages with many spelling mistakes? This might help against all
>>> those creative ways of spelling out what the spammer wants to sell.
>>
>>
>> I don't know that anyone has wor
Loren Wilton wrote:
I'm wondering, would it be useful to have a plugin that penalizes messages
with many spelling mistakes? This might help against all those creative
ways of spelling out what the spammer wants to sell.
I don't know that anyone has worked on specifically what you are thinking
of,
> > Is there any evidence that local.cf is getting read at all?
>
> Good question. Where do I look for such evidence? Logs are hardly
> revealing.
if you are running postfix,amavisd-new and spamassassin chrooted be sure to not
only change /etc/mail/spamassassin/local.cf but also
/etc/mail/sp
For the archives,
On Mon, 2004-10-04 at 09:32, Maurice Lucas wrote:
I use spamd and spamc with SA3.0 in a sitewide configuration.
A few percentage of all my connections keeps the spamc call in memory.
After a 500 connections I have 23 times "/usr/local/bin/spamc -c -u spamd"
in my ps list.
Does s
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Jeff Tucker writes:
> Michael Parker wrote:
> > On Tue, Oct 05, 2004 at 10:22:42AM -0700, Morris Jones wrote:
> >
> >>I watched a spamd child grow to 250MB yesterday on a single message. I
> >>have a suspicion that the memory usage growth is happeni
Michael Parker wrote:
On Tue, Oct 05, 2004 at 10:22:42AM -0700, Morris Jones wrote:
I watched a spamd child grow to 250MB yesterday on a single message. I
have a suspicion that the memory usage growth is happening on a whitelist
or bayes database maintenance event of some sort.
For folks that are
snowjack wrote:
Is there any evidence that local.cf is getting read at all?
Good question. Where do I look for such evidence? Logs are hardly
revealing.
--
Sahil Tandon
On Thursday, October 7, 2004, 7:37:32 AM, Chris Santerre wrote:
> http://www.packtpub.com/book/spamassassin
Congrats Chris and the other editors and authors of this new
book. And thanks to all SA folks too! :-)
Jeff C.
--
Jeff Chan
mailto:[EMAIL PROTECTED]
http://www.surbl.org/
56 matches
Mail list logo
