Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: Another crazy hot backup theory in a book

Re: Another crazy hot backup theory in a book

From: Jeremiah Wilton <jeremiah_at_wolfenet.com>
Date: Thu, 24 Sep 1998 14:38:14 -0700
Message-ID: <Pine.OSF.3.95.980924143228.6105A-100000@gonzo.wolfenet.com>


On Thu, 24 Sep 1998 satar_at_my-dejanews.com wrote:
>
> Well, it is true. When you are running a hot tablespace backup, any changes
> to that tablespace is in the redo log buffer (which is part of the SGA) until
> the END backup is issued.

The database continues to write to the datafiles. Any changed blocks are logged to the redo logs (yes, via the redolog buffer) in their entirety to allow for recoverability of split block reads by the O/S during copy.

The changes *are not* stored up somewhere in the SGA and reapplied when the backup is finished. Such an approach would not scale to highly transactional databases.

--
Jeremiah Wilton http://www.wolfenet.com/~jeremiah Received on Thu Sep 24 1998 - 16:38:14 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US