Bert, this period of inactivity can be caused by two situations (or both):

1) There are an enormous amount of files that must be processed and
filtering and checking takes that long ...for a 1000 clients
2) The schedules defined are setup to use DURATION and DURATION UNITS
together with RANDOMIZATION
    values that cause the scheduled events to be PENDING that long (check
the  Q EVE * * output for the nodes to see
    if it stays in "PENDING" state that long)

Another case may be that your MAX SESSIONS value does not allow all the
clients to backup at the same time and that only the
allowed number of clients do start immediately and the rest have to wait
for their RANDOMIZED turn.....or till the DURATION expires whereafter the
event will show a status of "MISSED"

Regards

Campbell Esplin
Excalibur Business Solutions
----------
> From: Bert Geebelen AT cem DOT be <Bert_Geebelen AT CEM DOT BE>
> From: Bert Geebelen AT cem DOT be <Bert_Geebelen AT CEM DOT BE>
> To: ADSM-L AT VM.MARIST DOT EDU
> Subject: Period of inactivity!
> Date: Thursday, February 05, 1998 2:08 PM
>
> Hi,
>
> I've got a sort of a problem! It's just that I don't know if it really is
a
>  problem, wether it's a design issue!
> Here it comes...
>
>    Circumstances:
>    We're using at this stage an ADSM-MVS environment (V2R1L10). The
clients
>     are at version 2 release 1
>    level 06. In between there is a WAN-backbone!
>    Facts!:
>    When we're scheduling the backup of about thousand clients, we're
>    experiencing a period of inactivity
>    between the actual connect from the client to the server and the
moment
>    at which the client starts
>    sending the changend files, wich need to be backupped.
>    This period of inactivity can take one to more than two hours!!!
(That's
>     the reason why I think he is not
>    only exchanging all kinds of information).
>    Questions!
>    For the moment we are investigating several things. Nevertheless, I'd
>    like to know if there are people
>    experiencing the same 'problems'! I also hope there's an explanation
for
>     this matter?!?
>
> Thanks anyway.
> Greetings,
> Bert Geebelen.