Veritas-bu

[Veritas-bu] VCS Clustered NetBackup Master Server

2008-01-29 16:28:10
Subject: [Veritas-bu] VCS Clustered NetBackup Master Server
From: selwyn <netbackup-forum AT backupcentral DOT com>
To: VERITAS-BU AT mailman.eng.auburn DOT edu
Date: Tue, 29 Jan 2008 13:08:30 -0800
We have our master server clustered with VCS 5.0 using Solaris 10.  It works 
very well for us.  

All of the media server information is now stored in the EMM database.  No 
matter where you backup up the catalog you will be backing up the device 
information for all of you media servers.

Recovering the catalog can be a little tricky.  We recover with a tape drive 
attached directly to the master server.  You will need to setup the filesystems 
on your recovery system to mirror those on your production server.  

We use a filesystem on shared storage which mounts on /opt/VRTSnbu/
for the netbackup catalog.  

I have copied for you below my notes from the last time I recovered the 
netbackup catalog in a DR test.  

Assign nbmaster name and ip address to an interface on dr-mars.

install netbackup on dr-mars ? master server name nbmaster

run  ../sg.build ?all ?ml 2
run ./sg.install

After installing netbackup the following symbolic links must be created.

ln ?s /opt/VRTSnbu/var/global  /usr/openv/var/global
ln ?s /opt/VRTSnb/db/staging /usr/openv/db/staging

These symbolic links are used to point to the database locations which are used 
in our clustered environment.  bprecover will only recover to the original 
locations.

run next command with database up

nbdb_admin ?dba nbusql  

stop netbackup ? ?/usr/openv/bin/goodies/netbackup stop?
all netbackup processes must be stopped before proceeding?use bpps ?a to make 
certain all netbackup and media manager processes are stopped.  Run bp.kill_all 
to stop processes that will not stop.

delete all file in the directory /opt/VRTSnbu/db/data except for vxdbms.conf.

rm: remove EMM_DATA.db (yes/no)? y
rm: remove EMM_INDEX.db (yes/no)? y
rm: remove NBDB.db (yes/no)? y
rm: remove NBDB.log (yes/no)? y

run command nbdb_admin ?auto_start NONE ? prevents NBDB from starting

run nbdbms_start_stop start

create_nbdb  -drop ?data /opt/VRTSnbu/db/data ?index  /opt/VRTSnbu/db/data 
?tlog /opt/VRTSnbu/db/data 

If the above command fails ? run these commands:

create_nbdb ?drop

nbdb_admin ?dba nbusql

create_nbdb  -drop ?data /opt/VRTSnbu/db/data ?index  /opt/VRTSnbu/db/data 
?tlog /opt/VRTSnbu/db/data 


netbackup stop

netbackup start

tpext ?loadEMM

cd /usr/openv
Run /usr/openv/volmgr/bin/driver/sg.build
use tpconfig to add robot and entry for drive 1.

  -- determine using robot.pl
tpconfig ?add ?drive 0 ?type hcart  ?path /dev/rmt/?cbn  -robot 0 ?robtype TLD \
-robdrnum 1 (make certain this is the first drive in the robot)  -- determine 
drive device by correlating information from drive.pl and robot.pl.

start ltid
vmupdate ?rn 0 ?rt TLD

start ltid

locate correct catalog file  -- Catalog_backup_Vault_1_1198419629_UBAK  
December 23, 2007.

/usr/openv/DR/catalog/Catalog_backup_Vault_1_1198419629_UBAK

bprecover ?wizard

put correct bp.conf and vm.conf file back in place

after recovery stop and start netbackup

+----------------------------------------------------------------------
|This was sent by selwyn_schultz AT hermanmiller DOT com via Backup Central.
|Forward SPAM to abuse AT backupcentral DOT com.
+----------------------------------------------------------------------


_______________________________________________
Veritas-bu maillist  -  Veritas-bu AT mailman.eng.auburn DOT edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
<Prev in Thread] Current Thread [Next in Thread>
  • [Veritas-bu] VCS Clustered NetBackup Master Server, selwyn <=