[Mailman-Users] slightly OT: on not becoming a spam source

[David Newman]

One or more AOL subscribers is reporting posts to one of our mailing 
lists as spam. Every time there's a post to this list, both we and our 
upstream provider get a feedback report from AOL. This report provides 
the message-ID for the offending email, but not the recipient's address.

AOL claims I should be able to find the offended recipient(s) with just 
the message-ID, but I do not believe they are correct about that. 
grep'ing for the message-ID in maillog returns *all* of the ~250 
recipients for each post.

I would like to find which AOL user(s) are complaining and unsubscribe 
them.

One method would be to write a script that sends an individual email 
containing a unique ID to each subscriber, and see if AOL sends feedback 
complaint(s) for those.

Questions:

1. In putting together a script, do I need to include this header:

Return-Path: <mylist-bounces at example.org>

("mylist" and "example.org" are fact-free examples)

2. Is there some better way of isolating a given recipient if all I have 
from a spam report is a single message-ID?

This is on OpenBSD 4.2 running postfix 2.4.3 and mailman 2.1.10.

Many thanks!

dn