Re: SQL Loader Oracle 8 bug???

From: <abalbekov_at_usa.net>
Date: 1998/10/30
Message-ID: <71d8sj$m0t$1@nnrp1.dejanews.com>#1/1

I've tested this on my database (Oracle8.0.4.0.0, AIX 4.2). Started direct-load session, killed the process. Nothing happened. Table being loaded is available (no rows as it was before test), database successfully restarted.

Could you ask for the source for this alert?

Albert

--
Albert Balbekov
Senior Consultant
GRT Inc
Stamford
tel: (203)324-2400

In article <tOqZ1.1257$jE5.191738_at_client.news.psi.net>,
  "Gary Eckhardt" <Gary_Newsmail_at_dci-ltd.com> wrote:


> Hi All,

>

> I received this mail message from a business associate,

> and I've never heard anything about it.  Has anyone else?

> This is a great impact because we're currently migrating

> our DB to 8.0.4.1 and perform direct loads exclusively.

>

> Thanks for any information.

>

> ----- cut here ----- cut here ----- cut here ----- cut here -----

>

> Please be aware, and make other interested parties aware (developers)

> that a high impact alert has been issued for Oracle8.

>

> The bug is generic across all platforms and results is data dictionary

> corruption.  It is fixed in 8.0.5.

>

> The nature of the bug is that when a SQLLoader direct load is interrupted

> the data dictionary is corrupted with incorrect segment header information.

> Corruption of this sort would require a point in time recovery or a

> complete rebuild of the database.

>

> The workaround is to always use conventional loads or ensure that

> direct loads are not interrupted or killed.

>

>


-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own

Received on Fri Oct 30 1998 - 00:00:00 CST