-------  Response follows: ----------
If the window on a day isn't big enough, then I would take a look at doing an
If the window on a day isn't big enough, then I would take a look at doing an
"Incremental by Date" during the week, and schedule a regular incremental on
the weekend.  This has been discussed recently, but in a nutshell, it works
like this:

1.  First the big assumption - You have a complete full backup - the initial
one perhaps.
2.  An Incr by date is run.  This will back up only files that have date/time
stamps newer than the last full backup.  This is run daily (hopefully - the
more often it's run, the shorter the runs, obviously).
3.  On the weekend, a regular incremental is run.

What happens?  In step 2, only new files and files that have newer dates get
backed up.  There is no "clean up" for deleted files, etc., and files that
have a "different" date that is less than the date of the last backup are not
touched.  There are several exposures here.  But programs that play by the
rules should be well protected.  On the weekend, the regular incremental does
not back up these files again - instead it only gets any new files, and then
does all the clean up - delets files, etc.  Files with "different" dates, such
as installed software with compile dates of last year are backed up now.

How do you implement this?  Simply set up 2 separate schedules - one with a
weekday frequency, that does the incr by date, and a second that does the
regular incremental with a frequency of say every Saturday.  You only need to
start the schedular once on the client - the server will switch to the
appropriate schedule as necessary.

Hope this is of some help

Jerry Lawson
jlawson AT thehartford DOT com


________________________Forward Header________________________
Author: INTERNET.OWNERAD
Subject: Re: Re[2]: Different includes/excludes
02-20-97 12:59 PM

File item 2 original document name: MEMO 02/20/97  12:59
File item 2 document type: PCDATA

Text of note follows:

Uh, Jerry,,, I started this thread (at least this time), and the reason we
are looking to do this is that ADSM passing data to the mainframe is to
 slow relative to the amount of data that needs to be passed and the
processing window.  The client just doesn't have time to send that data so,
they are trying to compromise and minimally get weekly backups of that
particular group of files.