Hi, Shae!
Brinson, Jason T via Mailman-users writes:
Posfix log (journalctl -u | grep demo-list):
Is that really all there is? Have you sent no test posts to demo-list@cc.gatech.edu?
Postfix_lmtp: demo-list@mailman.cc.gatech.edu lmtp:[127.0.0.1]:8024
and so on ... this is as expected since you have an "alternative Postfix configuration".
Postfix_vmap: demo-list@cc.gatech.edu demo-list@mailman.cc.gatech.edu
ditto
Received: from esa.hc4021-78.iphmx.com (esa.hc4021-78.iphmx.com [207.54.93.77]) by cc-postoffice-01.cc.gatech.edu (Postfix) with ESMTPS id AF674400A3C for <demo-list@mailman.cc.gatech.edu>; Wed, 22 Apr 2026 17:27:54 -0400 (EDT)
AB3E2400ACA appears to be something unrelated.
Which in the logs looked like this: [root@cc-postoffice-01 ~]# journalctl -u postfix | grep AF674400A3C Apr 22 17:27:54 cc-postoffice-01 postfix/smtpd[677096]: AF674400A3C: client=esa.hc4021-78.iphmx.com[207.54.93.77] Apr 22 17:27:54 cc-postoffice-01 postfix/cleanup[676116]: AF674400A3C: message-id=<BN0PR07MB876638D9CD221355993FF318A02D2@BN0PR07MB8766.namprd07.prod.outlook.com> Apr 22 17:27:54 cc-postoffice-01 postfix/qmgr[523389]: AF674400A3C: from=<shae@cc.gatech.edu>, size=14593, nrcpt=1 (queue active) Apr 22 17:27:54 cc-postoffice-01 postfix/lmtp[677032]: AF674400A3C: to=<demo-list@mailman.cc.gatech.edu>, relay=127.0.0.1[127.0.0.1]:8024, delay=0.3, delays=0.28/0/0.01/0.01, dsn=2.0.0, status=sent (250 Ok)
This is weird -- if the RCPT TO that Postfix sees is different from the "to=<...>" in the last line, normally there would be an "orig_to=<...>" clause. Oh ...
% host mailman.cc.gatech.edu mailman.cc.gatech.edu has address 130.207.7.32 % host cc.gatech.edu cc.gatech.edu has address 130.207.7.104 cc.gatech.edu mail is handled by 0 cc-gatech-edu.mail.protection.outlook.com.
Who is "cc-postoffice-01" (mailman.cc.gatech.edu)? What does esa.hc4021-78.iphmx.com have to do with this (is that cc-gatech-edu.mail.protection.outlook,com?), and why isn't the previous hop cc.gatech.edu?
The "alternative Postfix configuration" assumes that the alternate doesn't exist in DNS. The idea is that the vmap rewrites "demo-list-join@cc.gatech.edu" to "demo-list-join@mailman.cc.gatech.edu", and the latter will be reprocessed from the beginning, which restarts the routing process from transport_maps (since mailman.cc.gatech.edu is not a virtual_alias_domain or virtual_mailbox_domain -- right?) I don't think it's *necessary* for mailman.cc.gatech.edu to be a figment of Postfix's imagination, but it would be easier to analyze if it was invisible to other nodes.
I don't know the internals of Postfix, but the lack of orig_to= suggests to me that somebody upstream of mailman.cc.gatech.edu is misconfigured and stripped out the "-join" based on the settings of 'recipient_delimiter' and 'propagate_unknown_extension' (assuming the MTA is Postfix, otherwise that MTA's similar settings).
Regards, Steve
-- GNU Mailman consultant (installation, migration, customization) Sirius Open Source https://www.siriusopensource.com/ Software systems consulting in Europe, North America, and Japan