I meant filtering spam directly at the MTA, but *MY* MTA, not the CDR node - i.e. soon as it sees a header that it doesn't like, close the connection or return an error to the sending MTA, optionally either blocking off the IP of the sender for some time or acting like a tarpit.
This would also work nicely against spam. I usually find that various spams all come from the same IP address. When I block these off at my firewall, the spam output drops for a while - until some other idiot picks up one of the "Put your computer to use" scams... In terms of being a tarpit, it should probably act as a fake relay to that IP for the remainder of the time period and fake acceptance of mail to any receipient, but all emails go straight to /dev/null, very slowly - so that it doesn't eat up too much bandwith, but enough to be a pain for the sender. Regarding inchoate and this mailing list, yes blocking off the CDR node from sending the spam won't work as the mailing list software would mark your address as bad and after a while unsubscribe you. Perhaps in that case the MTA could just accept the mail, but route it quietly to /dev/null. Weeding out inchoate is as easy as looking for a small sized body that includes not much beyond a link to slashdot and a few other places, regardless of how often he changes his email address. True he may resort to other forms of spam, but then that puts the burden on him to adapt. I doubt he would - repeated attempts to get him to put information in as to why we should bother to follow his urls has always failed. On Mon, 6 May 2002 [EMAIL PROTECTED] wrote: > No it would not. It would be a disaster.
