Re: block chg tracking

From: Mladen Gogala <mgogala_at_yahoo.com>
Date: Fri, 13 Feb 2015 19:50:04 -0500
Message-ID: <54DE9BBC.7020900_at_yahoo.com>

Ah yes, the good, old merge backup. However, it's cheating. One does not do "recover datafile" if the data file becomes corrupt, one does "switch datafile to copy" and then applies logs. Yes, I agree, that's the way of restoring incremental backups only. In addition to that, this will require a fairly large amount of storage. Device type SBT is still cheaper when it comes to storage.

On 02/13/2015 03:17 PM, Seth Miller wrote:
> http://docs.oracle.com/database/121/ADMQS/backrest.htm#ADMQS0932
>
> Seth Miller
>
> On Fri, Feb 13, 2015 at 12:55 PM, Mladen Gogala
> <dmarc-noreply_at_freelists.org <mailto:dmarc-noreply_at_freelists.org>> wrote:
>
> On 02/13/2015 01:48 PM, Seth Miller wrote:
>
> If RTO is an issue how about a database backup that doesn't
> require a level 0 restore, only the the level 1's and the
> archive and redo logs.
>
> In fact, this is the default backup strategy for 12c.
>
> Seth Miller
>
>
> Huh? You mean that in 12c it is possible to restore incremental
> backup, without restoring the last full? Do you have a link to the
> place where I could read about it? That looks mighty strange.
>
> --
> Mladen Gogala
> Oracle DBA
> http://mgogala.freehostia.com
>
> --
> http://www.freelists.org/webpage/oracle-l
>
>
>

-- 
Mladen Gogala
Oracle DBA
http://mgogala.freehostia.com


--
http://www.freelists.org/webpage/oracle-l

Received on Sat Feb 14 2015 - 01:50:04 CET