Re: RMAN backups - physical standby

  • From: Peter McLarty <peter.mclarty@xxxxxxxxxxxxxxxxxx>
  • To: ahbaid@xxxxxxx
  • Date: Wed, 02 Jul 2008 09:28:42 +1000

Hi Ahbaid
I think you might be misunderstanding some issues with the physical
standby setup and how corruption is likely not to go unnoticed
Firstly you ship archive logs from your primary to your standby, this
means that if that file is corrupt then you would have had a corrupt
file when you did your backups on the primary. 
next the standby is applying those archive logs effectively determining
if they are corrupt when they are applied to your standby database.
Now RMAN will throw errors if it encounters corruption when backing up,
therefore if your DB is in some way corrupt then you can be reasonably
assured that RMAN will tell you.
Finally you could from time to time open your standby and do an export
this has been notoriously good for telling you if you have block
problems in the past



After that, it is pretty much out of your ability to 

On Tue, 2008-07-01 at 14:44 -0700, Ahbaid Gaffoor wrote:

> Anyone out there taking their RMAN backups off of a physical standby 
> instead of the primary?
> 
> I'm wondering how to mitigate the risk of taking a backup with some 
> corruption coming down the redo stream, getting applied, and only 
> knowing about it when I attempt to open my standby. Of course trying to 
> recover this from tape would mean the corruption is also backed up on tape.
> 
> The idea of shifting the backups to the standby is that I free those 
> resources on the primary.
> 
> thoughts?
> 
> Ahbaid
> --
> //www.freelists.org/webpage/oracle-l
> 
> 

Other related posts: