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: 8.1.6 goes mad until HD full

Re: 8.1.6 goes mad until HD full

From: Chuck Hamilton <chuck_hamilton_at_yahoo.com>
Date: Fri, 26 Oct 2001 09:22:27 -0400
Message-ID: <9rbo66$s74u3$1@ID-85580.news.dfncis.de>


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

"Hannes Erven" <h.e_at_gmx.at> wrote in message news:3bd95b7f$0$20542$5039e797_at_newsreader01.highway.telekom.at...

>

> My primary is concern is this: what makes oracle write archived
> redo 100 times more often as usual, and how do I find out? I can
> definitely say that it's not simply more connections, more updates,
> large imports or something like that.

Leaving tablespaces in backup mode.

Under normal conditions (i.e not in backup mode) only the data that's actually changed gets logged. When in backup mode any change causes the entire block to be written to the log. Get the picture? When backup mode is on you get archiving out the wazoo. This is why it's best when performing a hot backup without RMAN to do one tablespace at a time.

-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 7.0.3 for non-commercial use <http://www.pgp.com>

iQA/AwUBO9ljkGm5A4kkb4ZsEQJEGwCghgPTO2rwg3I77+lN7rSzmerZchgAnREy g7VQ+mC2RtI6+69t7kxWXNC8
=Wamr
-----END PGP SIGNATURE----- Received on Fri Oct 26 2001 - 08:22:27 CDT

Original text of this message

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