"Clive McDowell" wrote: > > > If it gets past this ACL, the message was too big to be scanned, or
> > > it's score was below SpamAssassin::Conf::required_hits.
> > You can use
> > > this data as the basis for further decisions.
> >
> > Hi Ian,
> >
> > thanks for pointing this out. I forgot to mention this. Of
> > course this is much more easy, but only works if you want to
> > deny based on SA's result. If you want to tag the mail for
> > further processing, you have to stick with the long version I
> > posted, because defers in warn acl statements pass the
> > control flow to the next acl statement.
>
> Patrick,
>
> exactly so. Our user base here is paranoid about losing genuine
> messages so we only reject on very high spam scores (>100). The rest
