RE: Bayes problem in 3.1.2?

2006-06-01 Thread Bowie Bailey
Kai Schaetzl wrote:
> Nels Lindquist wrote on Wed, 31 May 2006 10:04:54 -0600:
> 
> > I removed the /var/lib/spamassassin/3.001002
> > directory which was blank
> 
> It shouldn't exist after an install and if you ran sa-update it should
> have content. Maybe an sa-update that got interrupted?

I have two machines that I set up recently with SA 3.1.1.  On both of
them, the first sa-learn run failed.  I didn't check to see if it left
me with an empty directory, but it is likely.  The second run worked
fine in both cases.

So these issues with empty directories could come from people who ran
sa-learn and either didn't notice the failure, or didn't continue
working with it after it failed.

-- 
Bowie


Re: Bayes problem in 3.1.2?

2006-06-01 Thread Kai Schaetzl
Nels Lindquist wrote on Wed, 31 May 2006 10:04:54 -0600:

> I removed the /var/lib/spamassassin/3.001002 
> directory which was blank

It shouldn't exist after an install and if you ran sa-update it should 
have content. Maybe an sa-update that got interrupted?

Kai

-- 
Kai Schätzl, Berlin, Germany
Get your web at Conactive Internet Services: http://www.conactive.com





Re: Bayes problem in 3.1.2?

2006-05-31 Thread Nels Lindquist
On 31 May 2006 at 11:50, Theo Van Dinter wrote:

> On Wed, May 31, 2006 at 09:46:52AM -0600, Nels Lindquist wrote:
> > I tried upgrading from 3.1.1 to 3.1.2 today, but "spamassassin --
> > lint" turned up the following errors:
> > 
> > [766] warn: config: warning: score set for non-existent rule BAYES_80
> [...]
> > 3.1.1).  Is there some larger issue here?  "Non-existent rule..." 
> > doesn't fill me with confidence.
> 
> Possibly, but we can't know without seeing the rest of your debug output.  I'd
> guess that either the install didn't put all the default rule files in place,
> or a faulty sa-update run.

It was the latter--I should have mentioned that I checked for 
/usr/share/spamassassin/50_scores.cf and the BAYES_* scores were 
definitely set there.  I removed the /var/lib/spamassassin/3.001002 
directory which was blank, and now everything seems to be working 
fine.

Thanks for your rapid assistance!


Nels Lindquist <*>
Information Systems Manager
Morningstar Air Express Inc.



Re: Bayes problem in 3.1.2?

2006-05-31 Thread Theo Van Dinter
On Wed, May 31, 2006 at 09:46:52AM -0600, Nels Lindquist wrote:
> I tried upgrading from 3.1.1 to 3.1.2 today, but "spamassassin --
> lint" turned up the following errors:
> 
> [766] warn: config: warning: score set for non-existent rule BAYES_80
[...]
> 3.1.1).  Is there some larger issue here?  "Non-existent rule..." 
> doesn't fill me with confidence.

Possibly, but we can't know without seeing the rest of your debug output.  I'd
guess that either the install didn't put all the default rule files in place,
or a faulty sa-update run.

-- 
Randomly Generated Tagline:
"When a man steals your wife, there is no better revenge than to let him
 keep her."  - Sacha Guitry


pgp6XF3C5p7pS.pgp
Description: PGP signature


Bayes problem in 3.1.2?

2006-05-31 Thread Nels Lindquist
Hi there.

I tried upgrading from 3.1.1 to 3.1.2 today, but "spamassassin --
lint" turned up the following errors:

[766] warn: config: warning: score set for non-existent rule BAYES_80
[766] warn: config: warning: score set for non-existent rule BAYES_95
[766] warn: config: warning: score set for non-existent rule BAYES_00
[766] warn: config: warning: score set for non-existent rule BAYES_40
[766] warn: config: warning: score set for non-existent rule BAYES_50
[766] warn: config: warning: score set for non-existent rule BAYES_99
[766] warn: config: warning: score set for non-existent rule BAYES_60
[766] warn: config: warning: score set for non-existent rule BAYES_20
[766] warn: config: warning: score set for non-existent rule BAYES_05

I override the default bayes scores in a local configuration file--it 
seems to be choking on my score overrides, though I can't see 
anything obviously wrong with the syntax (which was working fine in 
3.1.1).  Is there some larger issue here?  "Non-existent rule..." 
doesn't fill me with confidence.


Nels Lindquist <*>
Information Systems Manager
Morningstar Air Express Inc.