On 17 Jun 2014, at 20:54, k...@rice.edu wrote:
On Tue, Jun 17, 2014 at 08:46:51PM +0100, Sabahattin Gucukoglu wrote:
>> On 17 Jun 2014, at 20:00, Tom Hendrikx <t...@whyscream.net> wrote:
>>> On 17-06-14 19:46, Sabahattin Gucukoglu wrote:
>>>> [06/17/2014 16:08:11] Burton-Bayesian Probability: 0.500000
>>>> Samples: 27 11933: [06/17/2014 16:08:11] no factors specified;
>>>> using default 11933: [06/17/2014 16:08:11] Result Confidence: 0.50 
>>>> 11933: [06/17/2014 16:08:11] _ds_setall_spamrecords() failed
>>> 
>>> This isdicates that something went wrong while writing to the css
>>> file. I don't know what might be the problem with the css stuff, as I
>>> use database backends only.
>>> 
>>> Could you try switching to the mysql or even the sqlite backend?
>>> Even if you fix this issue, the hash driver has other longstanding
>>> issues, that probably won't get fixed soon.
>> 
>> Ah, thanks.  I certainly took well to the sqlite3 driver, and it does happen 
>> that the machine has MariaDB on it.  Would that do?
>> 
>> However the reason I selected hash was sbph+markov.  That’s not going to be 
>> supported with sqlite3/mysql, is it?  I *suppose* I could live with this. :)
> 
> I believe they are in the current release. We are using PostgreSQL with 
> osb+markov
> now. By the way, I would recommend osb over sbph. It performs less DB work 
> and is
> generally more accurate.

I just wanted to say that this configuration works, using sqlite3.  Thanks!  
Evidently, the hash driver needs obsolescence.

Cheers,
Sabahattin


------------------------------------------------------------------------------
HPCC Systems Open Source Big Data Platform from LexisNexis Risk Solutions
Find What Matters Most in Your Big Data with HPCC Systems
Open Source. Fast. Scalable. Simple. Ideal for Dirty Data.
Leverages Graph Analysis for Fast Processing & Easy Data Exploration
http://p.sf.net/sfu/hpccsystems
_______________________________________________
Dspam-user mailing list
Dspam-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/dspam-user

Reply via email to