If populating the lists is not a problem, I think the easiest way to
create a one time set of 170 list headers is to use what Paul suggests.

That is what we do on our unix box. I use a template for the list headers;
a single file of just the mailing list names (this is also used by another
script that takes the file and creates all of the mail aliases); and a
script that just creates a newlist.sh file with mailx commands. I use
"listname" as keyword in the template for the script to insert each unique
listname (in the Title and Notebook=) so I don't have to manually edit.
This works well.

If you are on unix, I'd be happy to send you the template I use and
the script I run against it -- it assumes the same list owner= lines on
all lists however. Contact me privately.

I hope this is helpful in some way.

Sincerely, Trish

On Thu, 20 Jul 2006, Paul Russell wrote:

> On 7/20/2006 18:47, Harris, Jason (DIS) wrote:
>> I bet curl or wget would work, maybe.   Script generated header + mailx
>> and curl/wget + wa.cgi methods seem about the same amount of work.
>>
>
> We create thousands of lists each semester with automated processes that use
> shell and/or perl scripts. In general, a cron job runs a script that reads an
> input data file, creates list header files using a standard template, then
> submits a PUT job by email for each list to be created. The specific LISTSERV
> postmaster address that is used on these PUT jobs, and on the subsequent ADD
> jobs which populate these lists, leads to /dev/null, so we do not have a
> mailbox overflowing with job output. Later in the day, another cron job runs
> another script to verify that the required lists, directories, and aliases
> were created.
>
> --
> Paul Russell, Senior Systems Administrator
> OIT Messaging Services Team
> University of Notre Dame
> [log in to unmask]
>

-----
Trish Forrest
Queen's University