You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@spamassassin.apache.org by "Steven W. Orr" <st...@syslang.net> on 2005/04/29 17:56:37 UTC

Question about how to invoke spamd for SQL.

I'm sure that others have already figured this one out.

What options should spamd and spamass-milter be started with in a MySQL 
environment?

My goal is to use SQL per-user configs, and also SQL autowhitelists and 
SQL Bayes data. I run a server here at home with (effectively) two users, 
myself and my wife. (The reason for using MySQL is partially for education 
and partially because spamass-milter sort of requires it to avoid use of 
sitewide bayesian data.)

For SA, I see -q and -Q. It says that if I use either of those then I 
*have* to use -x.

For spamass-milter I see the -u option but I'm not confident of the 
interplay of the SA options. I'd like to hear what other people have 
already worked out.

TIA

-- 
Time flies like the wind. Fruit flies like a banana. Stranger things have  .0.
happened but none stranger than this. Does your driver's license say Organ ..0
Donor?Black holes are where God divided by zero. Listen to me! We are all- 000
individuals! What if this weren't a hypothetical question?
steveo at syslang.net

Re: Question about how to invoke spamd for SQL.

Posted by Michael Parker <pa...@pobox.com>.
On Fri, Apr 29, 2005 at 11:56:37AM -0400, Steven W. Orr wrote:
> I'm sure that others have already figured this one out.
> 

Yep, check out this presentation (mostly the notes, which I tried to
write in such a way that would make it easy to follow along):
http://people.apache.org/~parker/presentations/index.html

> What options should spamd and spamass-milter be started with in a MySQL 
> environment?
> 
> My goal is to use SQL per-user configs, and also SQL autowhitelists and 
> SQL Bayes data. I run a server here at home with (effectively) two users, 
> myself and my wife. (The reason for using MySQL is partially for education 
> and partially because spamass-milter sort of requires it to avoid use of 
> sitewide bayesian data.)

And because it's faster :)

> For SA, I see -q and -Q. It says that if I use either of those then I 
> *have* to use -x.

Since you're going to be putting everything in SQL, then you'll want
to use -q, you'll probably also want to create a user (ie spamd) and
run spamd with -u spamd.  This will cause spamd to run with reduced
system level privs, but that doesn't matter since all you need to talk
to is the SQL server.  Using -x, which is perhaps a misnamed option,
keeps the process from using on disk user prefs.

Read through the presentation, it should give you a good start.

> 
> For spamass-milter I see the -u option but I'm not confident of the 
> interplay of the SA options. I'd like to hear what other people have 
> already worked out.
> 

This part is up to you.  What you need to know is that in order for
everything to work correctly, you need to make sure that you are
sending the proper username to spamd, however that is done in
spamass-milter.

Michael