Xref: utzoo comp.mail.misc:3103 comp.mail.sendmail:1466 Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!uflorida!ziggy!usfvax2!tct!chip From: chip@tct.uucp (Chip Salzenberg) Newsgroups: comp.mail.misc,comp.mail.sendmail Subject: Re: Bouncing mail for expired users Message-ID: <2606895A.6CF7@tct.uucp> Date: 20 Mar 90 19:49:46 GMT References: <25F83AB4.290E@tct.uucp> <1990Mar15.112945.5329@newcastle.ac.uk> Organization: ComDev/TCT, Sarasota, FL Lines: 25 According to C.R.Ritson@newcastle.ac.uk (Chris Ritson): >Surely the right thing to do is to ensure that when YOU bounce mail, >YOU fill in the headers in the message going out, so that if this >message is bounced by a less intelligent mailer, then it ends up >(depending on your preferences) in the waste bin, or with the >postmaster. The important thing is that it must NOT bounce if it does >come back to you. I quite agree that good E-Mail neighbors generate bounce messages that can be easily recognized as such. However, in the context of a Deliver 2.0 delivery file like the one I posted a few days ago, it is assumed that there is some Smart Mailer out there who will recognize a non-zero exit status from Deliver and generate the bounce message. Smail 2.5 typically fulfills the Smart Mailer role, and its bounce messages always look like they came from the user MAILER-DAEMON@host. My partial bounce-loop solution posted in a previous article was *intended* as a partial solution for use in concert with Smail 2.5 or some equivalently "smart" mailer. "Life *is* pain, highness. Anyone who says otherwise is selling something." -- Chip Salzenberg at ComDev/TCT , "The Usenet, in a very real sense, does not exist."