On 11/09/2011 15:58, John Hardin wrote: >> Email addresses that are actually used conform, typically, to a >> fairly constrained set of regexps (but not a constrained list of >> valid addresses...) > > Ah. Let me throw another idea your way, then: milter-regex. It would > allow you to validate recipient addresses against those regexes and > reject at SMTP-time if they don't match. Then the catch-all would only > apply to the ones that match the regex but aren't at the moment actual > mailboxes. Would reducing the inbound volume that way be acceptable?
That's an interesting idea... I'd have to work out how to do that from Postfix - but it would definitely cut the volume of spam I have to process. > I have another data point that may help you figure that out. When you > sent this reply, I assume you sent it to the list and me, either both > as TO addresses or one as a CC address. I received two separate > messages with no mention of the other recipient. It seems postfix on > ingwaz is splitting one message into multiples on outbound as well. Whoops, I confused things. I'd accidentally replied to you directly - then, having pressed "send" - realised... so sent an identical message to the list. >> It still seems odd, to me, that the envelope address can't necessarily >> influence the spamassassin score... The envelope address, I suspect, is >> more relevant than any other address found in the headers. > It certainly can if the MTA puts the envelope TO address into the > headers and if there's a rule looking for it. So, how do I get Postfix, my MTA, to put the envelope address into the headers?
