>(before SET LIST):
>
> RDR FILE 0126 SENT FROM SMTP2    PUN WAS 0446 RECS 0007 CPY  001 A NOHOLD
> NOKEEP
> 15 Jan 1903 08:50:09 Processing file 0126 from SMTP2@MSACVM
> 15 Jan 1903 08:50:09 -> BSMTP from [log in to unmask]
> 15 Jan 1903 08:50:09 -> 1 recipient (1 local).
> 15 Jan 1903 08:50:09 -> Passed to LISTTEST for delivery.

This is without the VM userid, right? It just means you have the global list exchange feature (GLX) activated in LMail, which you probably should not do at this point, at least not until everything else works and you have updated LMail to a version that does not think it is running in 1903 ;-) If this is with the VM userid, I do not understand what makes LMail forward the message to LISTTEST instead of just delivering it.

> RDR FILE 0128 SENT FROM LISTTEST PUN WAS 0076 RECS 0009 CPY  001 A NOHOLD
> NOKEEP
> 15 Jan 1903 08:50:09 Processing file 0128 from LISTTEST@MSACVM
> 15 Jan 1903 08:50:09 -> BSMTP from [log in to unmask]
> 15 Jan 1903 08:50:09 -> Message has "no forward" (L-NF) flag.
> 15 Jan 1903 08:50:09 -> 1 recipient (1 local).
> 15 Jan 1903 08:50:09 -> Sent delivery error for 1 recipient.

This is because the GLX does not work in LISTTEST, presumably because it is not configured or the PEERS NAMES entry is not compatible with it or whatever. LISTTEST simply sends the message back to LMail so that it can generate the delivery error it would have sent had GLX not been activated in LMail.

Anyway, no VM userid, no SET LIST, the mailbox is by definition unknown and there are no reasons for this to do anything that issue a delivery error. This is normal, if done in an exotic way.

> and another (after SET LIST):
> 
> RDR FILE 0138 SENT FROM SMTP2    PUN WAS 0450 RECS 0007 CPY  001 A NOHOLD
> NOKEEP
> 15 Jan 1903 08:54:33 Processing file 0138 from SMTP2@MSACVM
> 15 Jan 1903 08:54:33 -> BSMTP from [log in to unmask]
> 15 Jan 1903 08:54:33 -> 1 recipient (1 local).
> 15 Jan 1903 08:54:33 -> Delivered to TEST.

That looks good. What does LISTTEST say then? Anyway, it is the correct setting and behaviour.

> Date:         Wed, 15 Jan 2003 08:54:34 -0800
> From:         [log in to unmask]
> Subject:      Undelivered mail
> To:           [log in to unmask]
> X-Report-Type: Nondelivery; boundary="> Error description:"
> 
> --> Error description:
> Error-For:  [log in to unmask]
> Error-Code: 3
> Error-Text: No such list.
> 
> Error-End:  One error reported.

LISTSERV will do that if you manage to delete the list between the time when the message was enqueued to LISTSERV and the time LISTSERV actually got to it. This is very difficult to do on VM (or even on other systems, but even more so on VM with a minidisk lacking concurrent write access). You would really have to engineer a series of commands and jobs that causes it to happen, which I doubt is the case here. You wouldn't happen to have your list as a zero-length SFS file, would you? A zero-length list file is not supported (there must be at least one header line) but, if you were to create one, it would be treated as a nonexistent list, because LISTSERV would be sorry if it tried to do anything with it.

If I had to take a bet, I would say that you are pointing to the wrong LISTSERV. The LISTSERV userid that sent this back has indeed no list called TEST, but it is not the LISTSERV that you meant to pass the message to. It is your production LISTSERV rather than your test LISTSERV, or vice-versa. It's great to be able to clone your hardware at no costs and with no performance degradations for testing purposes, but it does open up a whole new range of ways to shoot yourself in the foot when testing ;-)

  Eric