Re: [gentoo-server] spamassassin

Re: [gentoo-server] spamassassin

[Lance Lassetter]

This might be helpful to this thread:

http://www.apache.org/~parker/presentations/

Thanks,

Lance(stuNNed)

On Sun, 2005-03-13 at 22:50 -0600, Lance Lassetter wrote:
> in /etc/conf.d/spamd i have:
> 
> SPAMD_OPTS="-u spamfilter -r /etc/mail/spamassassin/spamd.pid"
> 
> where user spamfilter is mail filtering user set up in postfix.
> 
> lance(stuNNed)
> 
> On Sun, 2005-03-13 at 13:44 -0800, Ben Munat wrote:
> > Hmm, so are people using the -u option or letting spamd setuid (to "nobody" it appears...)?
> > 
> > Renzo Rosales wrote:
> > > Read the man page for spamd and see what parameter you can use to set
> > > which user spamd to run as. It may be a command-line entry so you
> > > might want to edit /etc/conf.d/spamd and see what options are being
> > > passed.
> > > 

--
gentoo-server@gentoo.org mailing list

Re: [gentoo-server] spamassassin

[D. Wokan]

Christoph Gysin wrote:

> Lance Lassetter wrote:
>
>> i'm interested if sa-learn can be done site-wide.  bayes does seem to
>> help a bit but on a system with 500 users it doesn't seem trivial.
>
>
> I have e special IMAP folder, where users can put in false-negatives 
> (spam that made it in your inbox). A cronjob scans that folder and 
> trains the sitewide bayes stored in a mysql-DB.
>
> Christoph

I do the same thing.  Granted I'm supporting a whopping 3 users, but I 
use a cron job to go through /home/*/.mail/SA-MissedSpam and 
/home/*/.mail/SA-FalsePositives nightly.  It even clears out the missed 
spam folder automatically so space doesn't get wasted on what wasn't 
wanted to begin with.

D. Wokan
--
gentoo-server@gentoo.org mailing list

Re: [gentoo-server] spamassassin

[Lance Lassetter]

yes, the @GLOBAL feature in SA's support in MySQL is quite nice.

I've just completed setting up AWL, bayes, and threshold support in MySQL
for SA.  The only thing is I'm using amavisd-new as well so the threshold
(set here to 5.5) is read from amavisd.conf and not from SQL.  However the
rest is, which is great!

If using amavisd-new and to check for this, run 'amavasd debug-sa' and
pipe a mail thought it, you will it connects to the db.

Lance(stuNNed)

> Christoph Gysin wrote:
>
>> Lance Lassetter wrote:
>>
>>> i'm interested if sa-learn can be done site-wide.  bayes does seem to
>>> help a bit but on a system with 500 users it doesn't seem trivial.
>>
>>
>> I have e special IMAP folder, where users can put in false-negatives
>> (spam that made it in your inbox). A cronjob scans that folder and
>> trains the sitewide bayes stored in a mysql-DB.
>>
>> Christoph
>
> I do the same thing.  Granted I'm supporting a whopping 3 users, but I
> use a cron job to go through /home/*/.mail/SA-MissedSpam and
> /home/*/.mail/SA-FalsePositives nightly.  It even clears out the missed
> spam folder automatically so space doesn't get wasted on what wasn't
> wanted to begin with.
>
> D. Wokan
> --
> gentoo-server@gentoo.org mailing list
>



--
gentoo-server@gentoo.org mailing list

Re: [gentoo-server] spamassassin

[Ben Munat]

I do this too... just wish spamassassin would hurry up and learn. My girlfriend gets the 
same spam over and over -- often several copies of the same one at a time -- everyday. 
Today I looked at the headers of a missed spam -- one of those "we've been trying to 
contact you about our great mortg(age (sic) opportunities" -- and it's score was 0.0!! 
Well, it does block some, so I guess that's something. It just seems like nothing will 
ever fix the spam problem.

b


D. Wokan wrote:
> I do the same thing.  Granted I'm supporting a whopping 3 users, but I 
> use a cron job to go through /home/*/.mail/SA-MissedSpam and 
> /home/*/.mail/SA-FalsePositives nightly.  It even clears out the missed 
> spam folder automatically so space doesn't get wasted on what wasn't 
> wanted to begin with.
> 
> D. Wokan
--
gentoo-server@gentoo.org mailing list

Re: [gentoo-server] spamassassin

[peter.gantner]

Quoth Ben Munat, on Saturday, the 26th of March:

> I do this too... just wish spamassassin would hurry up and learn. My 
> girlfriend gets the same spam over and over -- often several copies of the 
> same one at a time -- everyday. Today I looked at the headers of a missed 
> spam -- one of those "we've been trying to contact you about our great 
> mortg(age (sic) opportunities" -- and it's score was 0.0!! Well, it does 
> block some, so I guess that's something. It just seems like nothing will ever 
> fix the spam problem.
You could try training it with some spam from http://www.spamarchive.org

Worked nicely for me!

Greets,
 	Peter

-- 
"I do not think the way you think I think."
     -- Kai, last of the Brunnen G
--
gentoo-server@gentoo.org mailing list

[gentoo-server] Spamassassin

[Suranga Kasthuriarachchi]

[-- Attachment #1: Type: text/plain, Size: 319 bytes --]

Dear All,

I have installed the spamassassin, clamav and then qmailscanner.
qmailscanner working properly. but spamassassin is not working.
the content of the qmail-scanner-queue-version.txt showing only like bellow,

clamdscan: 0.88.4/1901. perlscan: 1.25st.

Please help me on this matter.

Suranga Kasthuriarachchi.

[-- Attachment #2: Type: text/html, Size: 351 bytes --]

[gentoo-server] SpamAssassin

[Wendall Cada]

I run several production servers on Hardened Gentoo, and have a concern.

The perl team has been awesome. I really love that packages are fairly
fresh compared to what is in cpan. However, there is one very important
package that's been ignored for far too long: SpamAssassin

http://bugs.gentoo.org/show_bug.cgi?id=177144

Why has it been waiting for a mips keyword for almost two years? No
reply from mips. Pinging the bug has no effect, as seen with comments
#5-#8. Is the package abandoned? Is there some political issue with
Gentoo that keeps abandoned packages from being reassigned? I would love
to offer my help, but lately those offers have been flat rejected by
Gentoo devs. I guess my lips don't pucker well enough. I would like to
have this package dealt with. There are some important bug fixes in
newer releases.

Wendall

Re: [gentoo-server] SpamAssassin

[David]

Wendall Cada wrote:
> I run several production servers on Hardened Gentoo, and have a concern.
>
> The perl team has been awesome. I really love that packages are fairly
> fresh compared to what is in cpan. However, there is one very important
> package that's been ignored for far too long: SpamAssassin
>
> http://bugs.gentoo.org/show_bug.cgi?id=177144
>
> Why has it been waiting for a mips keyword for almost two years? No
> reply from mips. Pinging the bug has no effect, as seen with comments
> #5-#8. Is the package abandoned? Is there some political issue with
> Gentoo that keeps abandoned packages from being reassigned? I would love
> to offer my help, but lately those offers have been flat rejected by
> Gentoo devs. I guess my lips don't pucker well enough. I would like to
> have this package dealt with. There are some important bug fixes in
> newer releases.
>
> Wendall
>
>
>   
Did you try #gentoo-perl

-- 
Powered by Gentoo GNU/Linux
http://linuxcrazy.com

Re: [gentoo-server] SpamAssassin

[Kerin Millar]

2009/4/7 Wendall Cada <wendallc@83864.com>:
> I run several production servers on Hardened Gentoo, and have a concern.
>
> The perl team has been awesome. I really love that packages are fairly
> fresh compared to what is in cpan. However, there is one very important
> package that's been ignored for far too long: SpamAssassin
>
> http://bugs.gentoo.org/show_bug.cgi?id=177144
>
> Why has it been waiting for a mips keyword for almost two years? No
> reply from mips. Pinging the bug has no effect, as seen with comments
> #5-#8. Is the package abandoned? Is there some political issue with
> Gentoo that keeps abandoned packages from being reassigned? I would love
> to offer my help, but lately those offers have been flat rejected by
> Gentoo devs. I guess my lips don't pucker well enough. I would like to
> have this package dealt with. There are some important bug fixes in
> newer releases.

I can appreciate your frustation. At this point, maybe you should try
bringing it to the attention of the QA team (qa@gentoo.org)? Resolving
this situation would presumably be in line with their first stated
goal, which is to "keep the tree in a state which benefits all of our
users and developers alike".

Another tactic would be to befriend a developer who has a vested
interest in the continued fitness of the package (i.e. is known to be
a user) and to ask for their assistance in moving things forwards.
Some developers are rather more receptive to contributions from users
than others, and you may chance upon one.

Cheers,

--Kerin

Re: [gentoo-server] SpamAssassin

[Wendall Cada]

Thanks Kerin. Sorry to show my frustration. Just my feeling toward parts
of the Gentoo project lately. I will ping QA about this, and try to find
a perl dev who might look at the package.

This package is pretty stable. There are just some important bug fixes
in newer releases that would be nice to have stabilized.

Wendall

Kerin Millar wrote:
> 2009/4/7 Wendall Cada <wendallc@83864.com>:
>   
>> I run several production servers on Hardened Gentoo, and have a concern.
>>
>> The perl team has been awesome. I really love that packages are fairly
>> fresh compared to what is in cpan. However, there is one very important
>> package that's been ignored for far too long: SpamAssassin
>>
>> http://bugs.gentoo.org/show_bug.cgi?id=177144
>>
>> Why has it been waiting for a mips keyword for almost two years? No
>> reply from mips. Pinging the bug has no effect, as seen with comments
>> #5-#8. Is the package abandoned? Is there some political issue with
>> Gentoo that keeps abandoned packages from being reassigned? I would love
>> to offer my help, but lately those offers have been flat rejected by
>> Gentoo devs. I guess my lips don't pucker well enough. I would like to
>> have this package dealt with. There are some important bug fixes in
>> newer releases.
>>     
>
> I can appreciate your frustation. At this point, maybe you should try
> bringing it to the attention of the QA team (qa@gentoo.org)? Resolving
> this situation would presumably be in line with their first stated
> goal, which is to "keep the tree in a state which benefits all of our
> users and developers alike".
>
> Another tactic would be to befriend a developer who has a vested
> interest in the continued fitness of the package (i.e. is known to be
> a user) and to ask for their assistance in moving things forwards.
> Some developers are rather more receptive to contributions from users
> than others, and you may chance upon one.
>
> Cheers,
>
> --Kerin
>
>