How do I use sa-learn for virtual users?

[Pat Lashley <pa...@volant.org>]

Sorry about the previous partial post - I'm not sure just which keys I
hit that made it post before I was ready...

I have SpamAssassin 2.63 installed and configured to handle Cyrus IMAP
virtual users.  The command-line parameters are:

    --daemonize
    --pidfile=/var/run/spamd.pid
    --username=mailnull
    --auto-whitelist
    --create-prefs
    --virtual-config-dir=/var/spool/spamassassin/%u
    --nouser-config

It's working fine and generating the expected user directories with
.spamassassin subdirs with the auto-whitelist and bayes files.  But
I haven't figured out how to train virtual users' bayes filters.
The sa-learn script doesn't seem to have any parameters that would
provide functionality similar to spamd's virtual-config-dir param;
and there doesn't seem to be any other simple training interface
available on the command-line.

So what's the best way to train the Bayes filters given per-user
verified-spam and verified-ham IMAP folders (also available as
directories of mail files) ?


Thanks,
-Pat

Re: How do I use sa-learn for virtual users?

[Martin Hepworth <ma...@solid-state-logic.com>]

Pat

I've got a perl script that assumes the imap dirs are on a different 
machine - it logs into the imap server, grabs the emails sa-learns them 
and then delete's the corrections.

You might be able to hack it to use for your purposes..let me know if 
you've like it..


--
Martin Hepworth
Snr Systems Administrator
Solid State Logic
Tel: +44 (0)1865 842300


Pat Lashley wrote:
> Sorry about the previous partial post - I'm not sure just which keys I
> hit that made it post before I was ready...
> 
> I have SpamAssassin 2.63 installed and configured to handle Cyrus IMAP
> virtual users.  The command-line parameters are:
> 
>     --daemonize
>     --pidfile=/var/run/spamd.pid
>     --username=mailnull
>     --auto-whitelist
>     --create-prefs
>     --virtual-config-dir=/var/spool/spamassassin/%u
>     --nouser-config
> 
> It's working fine and generating the expected user directories with
> .spamassassin subdirs with the auto-whitelist and bayes files.  But
> I haven't figured out how to train virtual users' bayes filters.
> The sa-learn script doesn't seem to have any parameters that would
> provide functionality similar to spamd's virtual-config-dir param;
> and there doesn't seem to be any other simple training interface
> available on the command-line.
> 
> So what's the best way to train the Bayes filters given per-user
> verified-spam and verified-ham IMAP folders (also available as
> directories of mail files) ?
> 
> 
> Thanks,
> -Pat
> 

**********************************************************************

This email and any files transmitted with it are confidential and
intended solely for the use of the individual or entity to whom they
are addressed. If you have received this email in error please notify
the system manager.

This footnote confirms that this email message has been swept
for the presence of computer viruses and is believed to be clean.

**********************************************************************