> >Even if it's not good for statistical purposes, what do you think about > >a new parameter "stop_on_hold=true/false" wich will stop any further > >(external, ressource intensive) tests if the hold value is already > >reached? > >I think some systems with a lot of traffic can then use more tests, but > >only if this is necessary. > > It's something we're thinking about. It would require a lot of work the > way Declude JunkMail is currently set up, but would be useful if it could > prevent resource intensive tests from being run unnecessarily.
I am not so sure about the usefulness of that. While some may like it, others will not. Here is why. What about increased weights get other actions? That would kind of throw different weight thresholds out the door. Example, if a message reaches 20, we hold. But if that same message reaches 45, it is deleted. And since weight tests are the last run (I think), Declude would not even know that at what weight hold was until the end. You get a message that fails enough minor tests to be held, so processing is stopped. You look at the message and say, well it failed NOPOSTMASTER, NOABUSE, REVDNS, MAILFROM and BADHEADERS and you so OK this one is ok. But what if it had adult content in it or something that would have been caught by another test to where you would have said no way delete. Just my tired opinion. John Tolmachoff MCSE, CSSA IT Manager, Network Engineer RelianceSoft, Inc. Fullerton, CA 92835 www.reliancesoft.com --- [This E-mail was scanned for viruses by Declude Virus (http://www.declude.com)] --- This E-mail came from the Declude.JunkMail mailing list. To unsubscribe, just send an E-mail to [EMAIL PROTECTED], and type "unsubscribe Declude.JunkMail". The archives can be found at http://www.mail-archive.com.
