Re: Silly question

Getting back to this, I decided to kill the two perl threads
so I could
run these at a more advantageous time.  Started to run the
script at a
more advantageous time and now the mysql process jacks the
processor and
each one takes 5 minutes to process!  I fear I may have
damaged something
in the db's but I don't know what.   Restarting the
processes didn't
help.
Since I'll never be able to catch up, is there a to just
start over with
the Bayes database?  We haven't rolled that part out to the
users yet so
it's no loss, only 2 weeks of monitored traffic.

-- 
Rick Zeman
Manager of Information Technology
Melwood Horticultural Training Center
301.599.4574 - HelpDesk
301.599.4560 - MyDesk
http://www.melwood.org

>>> David Sims <dpsimsdpsims.com> 9/27/2007
11:10:26 AM >>>
Hi,

  This 'length of time' per message can be effected by many
things...
Like
reporting to various organizations... Each report takes a
finite time
and
if there are network timeouts this can be long... but.... if
you run
process-quarantine in cron and have a big backlog, after the
hour passes
and the next instance of process-quarantine starts, it will
realize that
it is still running and do nothing.... Let it run for a
while and have a
look at the logging to see what it is doing.... Even though
it is taking
some time it is reducing the size of the backlog....

Dave
************************************************************
**************
On Thu, 27 Sep 2007, Rick Zeman wrote:

> Is it normal to take so long per message?
>
> 2007-09-27 09:49:41 Maia: [process-quarantine-sub]
Learned mail item
2335
> as non-spam
> 2007-09-27 09:49:54 Maia: [process-quarantine-sub]
Learned mail item
2340
> as non-spam
> 2007-09-27 09:50:10 Maia: [process-quarantine-sub]
Learned mail item
2369
> as non-spam
> 2007-09-27 09:50:51 Maia: [process-quarantine-sub]
Learned mail item
2379
> as non-spam
> 2007-09-27 09:51:22 Maia: [process-quarantine-sub]
Learned mail item
2383
> as non-spam
> 2007-09-27 09:51:28 Maia: [process-quarantine-sub]
Learned mail item
2396
> as non-spam
> 2007-09-27 09:51:39 Maia: [process-quarantine-sub]
Learned mail item
2404
> as non-spam
> 2007-09-27 09:51:52 Maia: [process-quarantine-sub]
Learned mail item
2418
> as non-spam
> 2007-09-27 09:51:58 Maia: [process-quarantine-sub]
Learned mail item
2425
> as non-spam
> 2007-09-27 09:52:05 Maia: [process-quarantine-sub]
Learned mail item
2439
> as non-spam
> 2007-09-27 09:52:10 Maia: [process-quarantine-sub]
Learned mail item
2454
> as non-spam
> 2007-09-27 09:52:15 Maia: [process-quarantine-sub]
Learned mail item
2471
> as non-spam
> 2007-09-27 09:52:20 Maia: [process-quarantine-sub]
Learned mail item
2516
> as non-spam
> 2007-09-27 09:52:43 Maia: [process-quarantine-sub]
Learned mail item
2547
> as non-spam
> 2007-09-27 09:53:06 Maia: [process-quarantine-sub]
Learned mail item
2556
> as non-spam
> 2007-09-27 09:53:18 Maia: [process-quarantine-sub]
Learned mail item
2557
> as non-spam
> 2007-09-27 09:53:29 Maia: [process-quarantine-sub]
Learned mail item
2561
> as non-spam
> 2007-09-27 09:53:36 Maia: [process-quarantine-sub]
Learned mail item
2573
> as non-spam
> 2007-09-27 09:53:54 Maia: [process-quarantine-sub]
Learned mail item
2588
> as non-spam
> 2007-09-27 09:54:03 Maia: [process-quarantine-sub]
Learned mail item
2630
> as non-spam
> 2007-09-27 09:54:03 Maia: [process-quarantine-sub] 20
non-spam items
> processed (20 learned)
>
>
> >>> Cory Coager <ccoagerutechproducts.com> 9/27/2007 8:41:46 AM
>>>
> You need to run process-quarantine.pl once in awhile. 
I run mine via
> crontab as the amavis user once an hour.  This will
process confirmed
> ham/spam and train the bayes database.
>
> ~Cory Coager
>
> Rick Zeman wrote:
> > When a user marks a message as spam (eg, it's a
false negative),
what
> > exactly is getting adjusted (and is this
irrespective of any config
> > settings?) and how?
> >
> > Reason I ask is that for a month we've been seeing
some of the same
> spams
> > to multiple uses with virtually the same scores
and marking them
> doesn't
> > seem to be adjusting their scores upwards at all.
> >
> > Bayes IS on, as is autolearn.  Any tuning tips I
possibly could have
> > overlooked?
> >
> > Tnx.
> >
> >
> >
> _______________________________________________
> Maia-users mailing list
> Maia-usersrenaissoft.com 
> http://www.renaissoft.com/mailman/listinfo/maia-users 
> _______________________________________________
> Maia-users mailing list
> Maia-usersrenaissoft.com 
> http://www.renaissoft.com/mailman/listinfo/maia-users 
>
_______________________________________________
Maia-users mailing list
Maia-usersrenaissoft.com
http://www.renaissoft.com/mailman/listinfo/maia-users