Should be no problem.

You would run "bprestore" within a script called by root's crontab
at the time of your choosing.

See Veritas Netbackup Datacentre 3.4 System Administrator's Guide for Unix
for the bprestore parameters.

We regularly run restores from our production Oracle server to our
development/fallback server to create clones of the database for
testing. In our case we use the rename facility to use different pathnames
from our production facility, but you may not need to.

The following is the guts of a (korn) shell script I use for my typical
restores. The variables are defined earlier in the script. ($restore
is the full path of the bprestore command):

$restore -B                     \
-l                              \
-L $log                         \
-R $rename                      \
-C $client                      \
-D $destclient                  \
-S $masterserver                \
-c $class                       \
-s ${start_date}                \
-e ${end_date}                  \
-w                              \
-f $filelist 

I had to do a little playing around in my vm.conf and bp.conf to get this
to work (e.g. both master servers need to be known by both sides), but
having done so, it works well.

As it happens I use "at" to run mine, on a one-off basis, but cron will be
equally fine. I have some logic in my script to check for a special
flag file I create to make sure it can't run by accident, but your need is
slightly different I guess.

Regards,

Mike Ellwood



> Hi, is it possible to schedule a restore of data from the 
> backup of one
> server to another standby server for the same time each week? 
> I am running
> Veritas datacentre 3.4.1 on HPUX but will be upgrading to 4.5 
> in the next
> couple of months.
> 
> Thanks in advance,
> 
> Steve.
> _______________________________________________
> Veritas-bu maillist  -  Veritas-bu AT mailman.eng.auburn DOT edu
> http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
>