See "bayes_file_mode" in the Mail::SpamAssassin::Conf man page.
Andreas Vogt wrote:
Is it possible to set acces rights for global bayes files in local.cf
(e.g. 660 or 666)?
---
This SF.Net email sponsored by: Free pre-built ASP.NET sit
On Sun, Aug 17, 2003 at 06:03:00AM +0300, Ryan Summers wrote:
> I noticed that the process seems to hang everytime it
> runs the RCVD_IN_OSIRUSOFT_COM test. Knowing little about
> this particular blacklist, I would guess that their rbl
> server is down.
Just another spammer burned by RBLs, think
On Sat, Aug 16, 2003 at 04:36:00PM +0300, Jason Staudenmayer wrote:
> This scored very low 3.8 any idea why?
Hard to say without complete headers. Maybe the spam was just
lucky and you weren't.
If you use Razor/DCC/Pyzor and RBL tests, perhaps the spam was
so new that no one had reported its fin
At 10:39 AM 8/16/03 -0700, [EMAIL PROTECTED] wrote:
I want to do this w/ DNSBL -- if a mail comes from an MTA that is listed on
2+ DNSBL lists, multiply the score by a bunch.
No, but you could use a meta rule and add more points..
Something like this:
meta LOCAL_MANY_RBLS ((RCVD_IN_DSBL + RCVD_IN
Hello Andrew,
Saturday, August 16, 2003, 1:29:54 PM, you wrote:
AC> I have multiple mailboxes in ~/mail, as do the other users on my
AC> server. ~/mail/spam/definitely_spam is an mbox that contains, well,
AC> spam. If I run:
AC> sa-learn --ham --mbox /home/someuser/mail/*
AC> it will autolearn
Appears to be working here doing a manual test. I think setting the
score to 0 disables the test, which you can do in your local.cf.
Ryan Moore
--
Perigee.net Corporation
704-849-8355 (sales)
704-849-8017 (tech)
www.perigee.net
~$ dig 2.0.0.127.relays.osirusoft.com
; <<>> DiG 9.2.1 <<>
Hello all:
I just built a new mail server using qmail + vpopmail and spamassassin to
filter out spam. The last mail server I built using this same configuration
didn't have any trouble with spamassassin. After I installed spamassassin
and the necessary perl modules, I ran some tests and notic
I think to do what you want, the sa team would have to store the bayes
"fragments" inside the database as well, otherwise you'd not get all
the benefits from virtualisation
Anyone know why the bayes "tokens" couldn't be stored inside blob
fields in a handy dandy dbi-based database?
Ricardo Kle
On Saturday, August 16, 2003 @ 5:34:59 PM [-0700], Theo Van Dinter wrote:
> On Sat, Aug 16, 2003 at 05:17:57PM -0700, Matt Thoene wrote:
>> Aug 16 16:00:32 ns1 spamd[14875]: Failed to run BAYES_50
>> SpamAssassin test, skipping: ^I(No write permission
>> to sdbm file at lib/Mail/SpamAssassin/Bayes
Hello together,
I use spamassassin 2.55 with bayes and spamass-milt on sendmail for
systemwide scanning.
I told my spamassassin to use systemwide bayes files in homedir of user
"nobody", which is used by spamc (-u nobody).
I also want users give the possibility to send false negatives or fal
Hi,
I apologize in advance if this should be posted to an alternate list...
I'm getting the below errors with v 2.60-cvs...
Aug 16 16:00:32 ns1 spamd[14875]: Failed to run BAYES_80 SpamAssassin test, skipping:
^I(No write permission
to sdbm file at lib/Mail/SpamAssassin/BayesStore.pm line 933,
On Sat, Aug 16, 2003 at 05:17:57PM -0700, Matt Thoene wrote:
> Aug 16 16:00:32 ns1 spamd[14875]: Failed to run BAYES_50 SpamAssassin test,
> skipping: ^I(No write permission
> to sdbm file at lib/Mail/SpamAssassin/BayesStore.pm line 933, line 42. )
>
> What sdbm file does this error refer too?
Hello Andrew,
On Saturday 16 August 2003 22:29, Andrew Clarke wrote:
> I have multiple mailboxes in ~/mail, as do the other users on my server.
> ~/mail/spam/definitely_spam is an mbox that contains, well, spam. If I
> run:
>
> sa-learn --ham --mbox /home/someuser/mail/*
>
> it will autolearn the
On Sat, Aug 16, 2003 at 01:29:54PM -0700, Andrew Clarke wrote:
> sa-learn --ham --mbox /home/someuser/mail/*
>
> it will autolearn the spam/definitely_spam mailbox as ham, which I
> obviously don't want. I can think of 2 workarounds:
>
> to get a good system going and then deploy it more widely.
At 19:57 16/08/2003, you wrote:
Having had a dig through PerMsgStatus::get_raw_body_text_array(), it spots
non-textual MIME sections and skips them.
Please could that be changed to a configuration option I can set a
run-time when the object is created?
I know all the arguments about large memor
I have multiple mailboxes in ~/mail, as do the other users on my server.
~/mail/spam/definitely_spam is an mbox that contains, well, spam. If I
run:
sa-learn --ham --mbox /home/someuser/mail/*
it will autolearn the spam/definitely_spam mailbox as ham, which I
obviously don't want. I can think o
On Sat, Aug 16, 2003 at 10:39:06AM -0700, [EMAIL PROTECTED] wrote:
> I want to do this w/ DNSBL -- if a mail comes from an MTA that is listed on
> 2+ DNSBL lists, multiply the score by a bunch.
You can't do that exactly, but you can do something like:
meta SOME_RULE ( dnsbl1 + dnsbl2 + ... dnsbln
Hi Guys,
Real creativity here. . .
Later--
Tim
-- Forwarded message --
Return-Path: <[EMAIL PROTECTED]>
Delivered-To: [EMAIL PROTECTED]
Received: from localhost (localhost.localdomain [127.0.0.1])
by localhost.localdomain (Postfix) with ESMTP id 2B9B56A72B
for <[E
Hi,
I'm using Windows 2000 as my desktop OS, together with Operas internal
mailer M2 as e-mail client. My ISP is running Spamassassin on Sun Solaris.
Is there a way I - with M2 or another email client - can train
Spamassassins bayesian filters for my mailbox there?
--
anders thoresson
Having had a dig through PerMsgStatus::get_raw_body_text_array(), it spots
non-textual MIME sections and skips them.
Please could that be changed to a configuration option I can set a run-time
when the object is created?
I know all the arguments about large memory footprints, etc, and am alread
I know that the calculations have changed in 2.60
to be more accurate, and that one result of this is that messages tend toward
0%, 50%, and 99% instead of the spread they used to have. I also realize
that the scores have not yet been adjusted to account for this.
That said, I seem to get
I want to do this w/ DNSBL -- if a mail comes from an MTA that is listed on
2+ DNSBL lists, multiply the score by a bunch.
Thanks!
---
This SF.Net email sponsored by: Free pre-built ASP.NET sites including
Data Reports, E-commerce, Portals, an
At 02:46 PM 8/16/2003 +0200, Jim Knuth wrote:
Hi,
how can I remove this part of header (between the stars)
X-Spam-Checker-Version: SpamAssassin 2.55 ***(1.174.2.19-2003-05-19-exp)***
Please do not remove that part of the header.. it's there on purpose and
it's intentional that there's no confi
This scored very low 3.8 any idea why?
--- Begin Message ---
Hi Dear friend,
I send you this email Because I find you sign some traffic exchange program
. Maybe you will be interested in some low
cost popup traffic .Yep , it's low cost and real traffic.
ADVERTISERS !! 1000 Unique PopUp only 1
At Sat Aug 16 14:23:34 2003, Paul Hutchings wrote:
> At present each message in the mbox file has all the headers listed, and all
> the spamassassin headers that were added when it first passed through - I'm
> cautious that I might be doing something dumb and training it that my own
> mail server
I use spamassassin on a relay that forwards to an Exchange 5.5 box. Users
stick any spam that made it undetected into a public folder.
I've connected to the exchange box using Netscapes mail client, so now I
have the contents of the public folder in mbox format. I'd like to use
sa-learn to train
At 02:01 PM 8/16/03 +0200, Eric Veltman wrote:
Hello everyone,
I recently installed SpamAssassin 2.50 from the SuSE 8.2 distribution CD.
It already filters out lots of spams so I don't have to read them, but there
are still several messages a day that are not detected as spam.
2.50 is HIGHLY vulne
At Sat Aug 16 13:01:09 2003, Eric Veltman wrote:
>
> Hello everyone,
>
> I recently installed SpamAssassin 2.50 from the SuSE 8.2 distribution CD.
> It already filters out lots of spams so I don't have to read them, but there
> are still several messages a day that are not detected as spam.
One
Hi,
how can I remove this part of header (between the stars)
X-Spam-Checker-Version: SpamAssassin 2.55 ***(1.174.2.19-2003-05-19-exp)***
thanks
cu Jim
---
This SF.Net email sponsored by: Free pre-built ASP.NET sites including
Data Reports
At 11:52 AM 8/16/03 +0100, Angel Gabriel wrote:
I would like to fine tune my SA install. The problem I have is that a
lot of spam isn't listed as spam. I have a filter that moves ANY message
that has a spam score, you know, when the * is added to the header. This
tends to help me filter all spam, b
Hello everyone,
I recently installed SpamAssassin 2.50 from the SuSE 8.2 distribution CD.
It already filters out lots of spams so I don't have to read them, but there
are still several messages a day that are not detected as spam.
For that purpose, I made a maildir folder where I collect all mess
At Sat Aug 16 09:52:15 2003, Simon Byrnand wrote:
> >
> > If you're on the Office 2003 beta programme, then you might consider
> > reporting the alleged change in behaviour (apparently, Outlook no
> > longer generates its own Message-ID, relying on the MTA to generate
> > one instead) as a possible
I would like to fine tune my SA install. The problem I have is that a
lot of spam isn't listed as spam. I have a filter that moves ANY message
that has a spam score, you know, when the * is added to the header. This
tends to help me filter all spam, but a lot of normal messages get
caught too, espe
> At Fri Aug 15 22:36:30 2003, Simon Byrnand wrote:
>
>> Exactly what version of Outlook or Outlook Express are you using ?
>>
>> As far as I can see the FORGED_MUA_OIMO rule is significantly broken and
>> falsely detects recent versions of Outlook/Outlook Express as being
>> "forged". I reported t
34 matches
Mail list logo
