Veritas-bu

Re: [Veritas-bu] Backing up cluster after failover

2009-06-12 11:50:25
Subject: Re: [Veritas-bu] Backing up cluster after failover
From: <judy_hinchcliffe AT administaff DOT com>
To: <VERITAS-BU AT MAILMAN.ENG.AUBURN DOT EDU>
Date: Fri, 12 Jun 2009 10:47:05 -0500
I don't think you are seeing the picture.

Go into your windows server and cluster admin.

Now look at your resources and find the virtual name of the server that owns 
that resource.

This is NOT a physical server name.

Example when you go into cluster Administaff Inc.
You see a folder called Groups
Open up that 
In there you will se some server names (this should be DIFFERENT then the 
physical server names.

Click on each one of those and find the drive you want to backup - say the E 
drive
So in groups you see 3 names
You click on each one of them and find the one that has the E drive.

Use that server name in your nebackup policy.

Reason is that when the resource fails over from physical serverA to physical 
serverB the virtual name the resource belongs to is still the same name.

-----Original Message-----
From: veritas-bu-bounces AT mailman.eng.auburn DOT edu 
[mailto:veritas-bu-bounces AT mailman.eng.auburn DOT edu] On Behalf Of 
smwoodcrafts
Sent: Friday, June 12, 2009 10:34 AM
To: VERITAS-BU AT MAILMAN.ENG.AUBURN DOT EDU
Subject: [Veritas-bu] Backing up cluster after failover



ewilts wrote:
> On Fri, Jun 12, 2009 at 9:48 AM, smwoodcrafts <netbackup-forum < at > 
> backupcentral.com (netbackup-forum < at > backupcentral.com)> wrote:
> 
> The client name set on the host is actually irrelevant - it's only used when 
> you want to do a restore.
> 
> On your Windows cluster, you want to make sure that you have 3 backup 
> policies - 1a, 1b, and vir1.  The policies for 1a & 1b would back up just the 
> OS - for example, C: and System_State:.  You could likely combine those 2 
> into a single policy if the drive letters are the same between the systems.  
> The vir1 policy would back up just the volumes that are floating between the 
> hosts - for example, E: & F:.  You have to make sure you don't try to back up 
> the data volumes in the OS policy or you will get a status 71. 
> 
> 
> 
> >     .../Ed 
> > 
> > Ed Wilts, RHCE, BCFP, BCSD, SCSP, SCSE 
> > ewilts < at > ewilts.org (ewilts < at > ewilts.org)
> 


This is the problem, the business owners don't care about the OS or local 
drives. They do an SQL back up and have it written on one of the floating 
drives. The only thing we are backing up is the backup files so I only need the 
one policy for the cluster (vir1). The problem is that the active node changed 
and the host properties are still pointing to the old active node. Not the new 
active node. In other words, I go to host properties for the cluster. The 
Client tab says 1a (old active) not 1b (new active).

+----------------------------------------------------------------------
|This was sent by dseymour AT ap DOT org 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