Dear all,
 
We are using the format suggested by Eric below but not using the word
'import' at the end of the quiet add statement.  Does this make a
difference?.  Sometimes (to reduce the size of the job) I have dummied the
names as in
 
[log in to unmask] x x
instead of
[log in to unmask] Sandy Barnes
 
as the name really does not show up in the messages except in the welcome.
With the names, the job to regenerate the list is 275K.  We are just going
to keep growing so I would like to to know how to get around this problem.
 
Peter asked why we want to regenerate the list.  We keep information about
our subscribers in an external database also.  We currently have to
maintain the information in the database and the list with parallel
commands (update the database, update the list).  Because we are careful,
this works ok however why do it twice?  What I would like to do is to
maintain the changes in the database and weekly just generate a new list.
Only one job to run each week for each list (we have 5 major ones) instead
of daily listserv jobs maintaining the lists.  Also that way our list
always corresponds to our database (for better or for worse!) and we don't
have any drift.
 
Thanks for your help!
Sandy Barnes
Seattle, WA
 
 
On Wed, 8 May 1996, Eric Thomas wrote:
 
> The "job  stream" really  refers to  the size of  the *commands*  sent to
> LISTSERV, as opposed to data carried  into //XYZ DD sections. So what you
> should be doing is:
>
> quiet add xyz-l dd=xyz import
> //xyz dd *
> user1
> user2
> ...
> /*
>
>   Eric
>