REPOST: Re: Database Extents - Export and Import

From: Niall Litchfield <n-litchfield_at_audit-commission.gov.uk>
Date: Thu, 24 Jan 2002 09:41:05 -0000
Message-ID: <7$--$%%%_$$$_$%-$$_at_news.noc.cabal.int>


"Frank van Bortel" <fbortel_at_home.nl> wrote in message news:3C4F11A9.4E7E4B5_at_home.nl...
> David Williams wrote:
> >
> > "David Williams" <david484_at_hotmail.com> wrote in message
> > news:c43100d67ba014d8128c7b3016ff0296.36554_at_mygate.mailgate.org...
> >
> > > I am trying to clean up an ongoing extent problem with Oracle. I would
like to
> > > export all of the tables, clean up the extent issues and then import
the
> > > tables. I have never gone through this process. Is there anybody out
there that
> > > can walk me through this process?
> > >
> > > Thank you,
> > >
> > > David Williams
> >
> > My 'Extent issues' are in the tens of thousands. Do you have any
> > suggestions
> > for
> > a simple cleanup.
> >
> > --
> > Posted via Mailgate.ORG Server - http://www.Mailgate.ORG

>

> export, use imp indexfile= to get the DDL, edit to reasonable extents,
> re-create objects, import using ignore=y.
> Next time, make sure to use compress=n while exporting.
>
> Simple, very effective - but time consuming!

I have 2 comments on this.

First you *might* find that you can export, recreate the tablespaces as locally managed with an appropriate extent size. Then the import will use these extent sizes. You may find though that you need several tablespaces each with differing extent sizes and that therefore you need to edit the ddl as outlined above.

second. if the export was done with compress=n you wouldn't normally have tens of thousands of extents. you'd have one massive one. Compress=y says through away existing extent information and create one big enough to hold the entire table. (In the caseof LMTs this extents sizing would get silently ignored).

--
Niall Litchfield
Oracle DBA
Audit Commission UK
*****************************************
Please include version and platform
and SQL where applicable
It makes life easier and increases the
likelihood of a good answer

******************************************

========= WAS CANCELLED BY =======:
From: "Niall Litchfield" <n-litchfield_at_audit-commission.gov.uk>
Control: cancel <3c4fd6b4$0$8510$ed9e5944_at_reading.news.pipex.net>
Subject: cmsg cancel <3c4fd6b4$0$8510$ed9e5944_at_reading.news.pipex.net>
Date: Mon, 28 Jan 2002 02:33:49 GMT
Message-ID: <cancel.3c4fd6b4$0$8510$ed9e5944_at_reading.news.pipex.net>
X-No-Archive: yes
Newsgroups: microsoft.test,alt.flame.niggers,comp.databases.oracle.tools
NNTP-Posting-Host: w088.z064003087.lax-ca.dsl.cnc.net 64.3.87.88
Lines: 1         
Path: news.uni-stuttgart.de!dns.phoenix-ag.de!newsfeed01.sul.t-online.de!t-online.de!fr.clara.net!heighliner.fr.clara.net!nerim.net!feed.ac-versailles.fr!out.nntp.be!propagator-SanJose!in.nntp.be!news-in-sanjose!sjc-feed.news.verio.net!sea-feed.news.verio.net!news.verio.net!msrnewsc1!cppssbbsa01.microsoft.com!tkmsftngp01!tkmsftngp04!u&n&a&c&anceller
Xref: news.uni-stuttgart.de control:40719302

This message was cancelled from within The Unacanceller's glorious new software, Lotus 1-2-3 For Rogue Cancellers.
Received on Thu Jan 24 2002 - 10:41:05 CET

Original text of this message