[TriLUG] mailing list server filtering setup questions

[Michael Hrivnak]

I just use their rules from SARE.  I am not aware of any false positives on my 
systems that were generated while using updated SA rules and SARE recommended 
rules.

It's worth noting that it is very valuable to train these filters with 
sa-learn before putting them into production.  At the very least, run a few 
thousand ham messages through.  Especially in the early stages, be careful to 
collect as many false negatives as you can and run them through sa-learn.

Especially in your case, this will be a lot easier if you keep bayes info in 
something like a mysql database.  That makes it easy to run sa-learn on the 
machine where email is stored and have the bayes data go into the live 
database being used by spamd on your relay machine.

Michael

On Wednesday 26 December 2007 3:12:14 pm Cristóbal Palmer wrote:
> On Dec 20, 2007 5:26 PM, Michael Hrivnak <mhrivnak at hrivnak.org> wrote:
> > Consider taking advantage of this: http://saupdates.openprotect.com/
>
> Thanks again so much for your feedback. Tell me: do you use their
> plugins too or just their rules? If you use their plugins, how did you
> vet them before enabling them?
>
> Cheers,
> --
> Cristóbal M. Palmer
> celebrating 15 years of sunsite/metalab/ibiblio:
> http://tinyurl.com/2o8hj4


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part.
URL: <http://www.trilug.org/pipermail/trilug/attachments/20071226/448fbb4a/attachment.pgp>