Dave,
I'm just learning about NetBackup, so bear with me. Onward...
There are scripts with specific names which you can place in
/usr/openv/netbackup/bin on both servers and clients. If these scripts
exist in the bin directory, they will be executed when certain events
occur.
On a UNIX server there can be
backup_notify (media server, executes each time a backup is
successfully written to media)
backup_exit_notify (master server, executes when an individual backup
has completed from the
perspective of the client, media manager and image db)
dbbackup_notify (master server, executes each time server
completes a catalog db backup)
diskfull_notify (media server, executes if media server
encounters a disk full condition when
writing a backup to a disk type storage unit)
restore_notify (media server, executes when media server
completes sending data to a
client during a restore)
session_notify (master server, executes at end of a backup
session if at least one scheduled
backup has completed, can be used for server-side
post-processing)
session_start_notify (master server, executes when scheduler
detects a backup due to run, can be
used for server side pre-processing)
userreq_notify (master server, executes when a request is made to
list files in a backup or
archive, or start a backup/archive/restore)
On a UNIX client there can be
bpstart_notify (executes when client starts a backup or archive,
can be used for client side
pre-processing, be careful of BPSTART_TIMEOUT global
parameter on master
server (def = 300 seconds))
bpend_notify (executes when client completes a backup or
archive operation, can be used for
client side post-processing, be careful of
BPEND_TIMEOUT global parameter on
master server (def = 300 seconds))
The *TIMEOUT values represent the time allotted by the server to receive a
completion status value back from the client for the execution of the
*notify script. The bpstart_notify and bpend_notify can be made class and
schedule specific by extending the names as follows:
bpstart_notify.<class>
bpstart_notify.<class>.<schedule>
bpend_notify.<class>
bpend_notify.<class>.<schedule>
I believe that one of the parameters delivered to bpend_notify is the
completion status of the backup. I think it's $5, so you can test that and
do what you need with it. I have to think that the other scripts also have
status variables floating around that you could test.
Most of this information is from the NetBackup System Administrator's Guide
under "Notify Scripts" in the "Managing NetBackup" chapter. I am currently
looking at a 3.1.1 version, so it's probably good for 3.2 also. Sample
templates for these scripts typically are found in the
/usr/openv/netbackup/bin/goodies directory. You modify them as needed then
place them in the bin directory (make sure they are executable). The same
client scripts are available for Windows, but the names vary. Not sure
about server scripts for Windows.
Jeff Dykzeul
Raytheon Electonic Systems
"Dave Brown" <dbrown AT gfs DOT com>@mailman.eng.auburn.edu on 08/01/2002
11:27:00
Sent by: veritas-bu-admin AT mailman.eng.auburn DOT edu
To: <veritas-bu AT mailman.eng.auburn DOT edu>
cc:
Subject: [Veritas-bu] success vs. failure notifications?
Does Veritas Netbackup 3.2 have a built in reporting tool that will give me
a daily report for how many backups started each day and of those, how many
were either successful or failed?
I would like a daily report when I come in each day that would say
something like last night 135 backup jobs started with 15 jobs failing or
incomplete.
I haven't been able to find anything in the admin. guide yet, but I thought
I had seen something like this in other environments.
Thanks!
Dave Brown
Open System Administrator
email: dbrown AT gfs DOT com
phone: 616-381-2297
pager: 1-888-614-6762
_______________________________________________
Veritas-bu maillist - Veritas-bu AT mailman.eng.auburn DOT edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
|
