[Mimedefang] Training SA when mail is not stored locally?

[Nels Lindquist]

On 5 Feb 2004 at 11:44, Joe Arnstein wrote:

> How would learning be affected on a machine that receives ONLY spam?
> For example, our secondary server receives a steady flow of garbage all
> day and night, and only gets good stuff when the primary one goes down.
> If all it ever gets is garbage, how will it know what is legitimate when
> it sees it?  Won't its learning curve be skewed such that it knows good
> spam and bad spam?  :)

I'd not bother doing initial bayes training on the secondary. 
Instead, once you've got a decent bayes system working on your 
primary, copy the bayes_* database files over to the secondary and 
enable bayes.  (Make sure bayes_auto_learn is still off, though).  If 
you periodically copy your primary bayes databases to your secondary 
MX, the system should work farily well.

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