Just to make sure befor I make a fool of myself yet
again...
I set up SA to process all mail handled by sendmail.
Tell SA to tag them
then set the filter on the clients to look for my tag
and to shove it where they want it.
--- A New Razor User <[EMAIL PROTECTED]> wrote:
>
> > What I have not been able to determin is:
> > 1 - What does razor do with an email that is spam?
>
> Nothing, by itself. You need to use razor-report.
> It analyses the email
> and determines an appropriate signature method to
> use and files those
> signatures with the cloudmark servers.
>
> > Does it tell spamassassin that its spam?
>
> Spam-Assassin makes use of the Razor code to get an
> analogue "level of
> Spamminess", but unless you reconfigure the
> SpamAssassin score, by itself, a
> hit on razor isn't enough to call an email spam with
> a default SpamAssassin
> configuration.
>
> > Does it tag it in some way?
>
> SpamAssassin does the tagging, as appropriate. The
> razor-client doesn't
> change a single byte of the emails. You don't
> invoke razor scans of inbound
> emails, SpamAssassin makes use of the razor-client
> code to do that as part
> of its many checks.
>
> Think of SpamAssassin as a swiss army knife with
> hundreds of different
> blades with tunable sharpness. I make
> Microsoft_executables always trigger
> a SPAM determination, for example, as I have a
> policy which forbids users to
> email naked executables through my server.
>
> > 2 - Can I set up a filter in my client to forward
> > spam to a local MTA account on the server for
> > report/revoke at a later date?
>
> Yes. I like doing this with procmail though, as
> it's easier to retain ALL
> of the headers in as intact a form as possible.
> Some UNIX clients offer
> "redirect" or "bounce" functionality which do a good
> job, though.
>
> One real headache with SpamAssassin/razor tools in
> general is that so few
> email clients can forward emails completely intact
> as they were received.
> Getting these emails intact is vital for the
> Bayesian statistical processing
> parts of the anti-spam strategies.
>
> I use a strategy as follows:
>
> I set my SA spam threshold to 7.5
> For a "normal" configuration, I file all SA scores
> of 6.0 to 7.99 into a
> user's "Spam-Possible" folder.
>
> SA of 8.0+ goes to an automated reporting address
> which sa-learn --spam's it
> and razor-reports it. I've monitored this closely
> over a period of six
> months, and not a single false positive arrived with
> that score profile
> (N=90000).
>
> For users more concerned with false positives for
> whom receiving marketing
> drivel from Spamazon/ebay is of vital importance, I
> have tagging only, and
> no auto-filing to Spam-Possible with SA Score >= 9
> for auto-reporting.
>
> =R=
>
>
__________________________________
Do you Yahoo!?
Yahoo! Mail - 50x more storage than other providers!
http://promotions.yahoo.com/new_mail
-------------------------------------------------------
This SF.Net email sponsored by Black Hat Briefings & Training.
Attend Black Hat Briefings & Training, Las Vegas July 24-29 -
digital self defense, top technical experts, no vendor pitches,
unmatched networking opportunities. Visit www.blackhat.com
_______________________________________________
Razor-users mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/razor-users
