At 10:54 PM -0500 12/20/01, David W. Tamkin is rumored to have typed:
You have to make sure that the target address is in a place that the bad mail transport will send back to you.
You are, of course, correct. Most of the time, the return will include either the header fields (I have our sendmail adding the X-Envelope-To: header field, which admittedly not all do) or the body, but sometimes there are those brain-damaged mailers like FirstClass that cause the problems posters to this list are seeing from exon.massart.edu - it's bouncing messages sent from the mailing list server not to the envelope sender, but to the From: header field (Yuck!), and isn't including _any_ of the message whatsoever; no header, no body, no nuthin'. Even though the address might be inferred from the non-standard bounce message it's returning (Yuck! again, and not a guarantee if there are aliases set up in the thing), since it's sending it to the wrong address in the first place there's nothing the Lists.RWTH-Aachen.DE machine can possibly do about auto-unsubscribing the address anyway. In this case the only solution seems to be a polite email to the SmartList listowner, and a...er...different kind of mail to the Massachusettes College of Art. To drag this message into a semblance of on-topic relevancy, it would be impossible for Smartlist to intepret the useless FirstClass 6.0 error message even if it _did_ come to a SmartList server (which it wouldn't, of course, except possibly in the case of a digest list), and so the listowner would need to manually unsubscribe the address anyway. But fortunately, that's a rare (albiet _really_ annoying) condition. Usually bounce messages contain either headers, or partial body, or both of the rejected messages, so there's usually something to track. Charlie (who's almost getting ticked off enough at massart.edu to start rejecting _all_ SMTP connections from 'em)