Those are all good cases and it's important to be able to find information
about them.  But are you saying that LISTSERV puts those in the
listname.autodel file?  I've never seen this before in the 6 years I've
been working with LISTSERV here.

I thought the listname.autodel file was to keep track of addresses
*subscribed* to the list but that are bouncing for some reason.  What I'm
seeing is addresses that have never been subscribed to the list, in the
listname.autodel file and the report to the owner saying these addresses
are being monitored.

On Fri, 5 Mar 2004, Francoise Becker wrote:

> On 5 Mar 2004 at 14:30, James Morrill <[log in to unmask]> wrote:
>
> > What I am worried about is LISTSERV adding e-mail addresses to the
> > autodel file that aren't subscribed to the list in the first place and
> > then the querries I get from those owners wondering why LISTSERV is
> > tracking those people.
>
> Whenever LISTSERV responds to a command that is specific to a list,
> or a post to a list, LISTSERV uses the bounce address for the list.
> To understand why, here are a few of the many scenarios in which you
> might know about bounces for addresses who are not subscribed to the
> list:
>
> Scenario 1: Subscriber with misconfigured email
>
> Adam Jones is subscribed to a Send=private list as [log in to unmask]
> Unfortunately, he misconfigured his email client on his laptop and
> mail from there looks like it's coming from [log in to unmask] He posts
> to the list, and the email is rejected because [log in to unmask] is not a
> subscriber. Rejection bounces because [log in to unmask] is not a valid
> address.  Back at his regular workstation he goes to check his email
> and sees that the post he sent earlier was not distributed. He writes
> an angry email to the list owners screaming that they are censoring
> him, etc. etc. Owner checks bounces and sees a bounce from
> [log in to unmask] and is able to calm subscriber down.
>
> Scenario 2: Subscriber with forwarded email address
>
> Beth Smith subscribed as [log in to unmask], but somewhere along the
> line, she started to have [log in to unmask] forward all mail to
> [log in to unmask] She doesn't check her mail for a month, and
> her hotmail account starts bouncing. Bounces are coming from
> hotmail.com even though she's not subscribed there. (With probing,
> LISTSERV will eventually find out which subscribed address is
> bouncing, but in the meantime, the bounces are all the hotmail
> address). List owner checks the list for the hotmail address, doesn't
> find it, and starts looking for variations -- searching for "smith"
> turns up "Beth Smith" <[log in to unmask]>... Owner says "Hmmm...", sends
> an email to [log in to unmask] to ask if she's been getting her list
> email and receives a bounce from [log in to unmask] Aha! Set
> that one to NOMAIL or delete.
>
> Scenario 3: Non-subscriber tries to subscribe w/ bad address
>
> Chuck Taylor ([log in to unmask]) wants to subscribe to the list.
> Goes to the web interface but enters address as [log in to unmask]
> Since the list is OF COURSE set up for subscription confirmation, the
> confirmation bounces. Bounce is registered in monitoring report. A
> few days later, Chuck writes to list owners and complains that he
> subscribed to the list but isn't getting any email and all his posts
> are rejected. Once again, the fact that the non-subscribed address
> bounce is logged saves the day for the list owner who can reconstruct
> what happened.
>
> These are just a few of many scenarios in which logging bouncing non-
> subscribers is useful.
>
> Francoise
>
>

   ====================================================================
 James Morrill  office: Hale 11, 785-532-4909 www-personal.ksu.edu/~james
             We see a world of wonder, with a holy fingerprint
     But we only know a sliver, of the love of God in it. - Hokus Pick
   ====================================================================