Re: check for (empty?) blocks which leads to ORA-8103 in a datafile?

From: Martin Berger <martin.a.berger_at_gmail.com>
Date: Tue, 8 Sep 2009 21:53:22 +0200
Message-Id: <73E68CA6-57E3-4C09-97F1-17F2D2EF8F40_at_gmail.com>



did not show any errors.

I assume, Oracle does not initiate any blocks within a datafile before it uses it.
In my particular case, oracle tries to get a block, but instead it's a valid block, it's empty.
The error in the logic is (I still guess) not to report it as 'broken' but as 'not used anymore by _this_ object' - which leads to ORA-8103. (In my idea, there should be the knowledge 'if it should ever have been part of an object, but now it's empty, something went terrible wrong. But this could leasd also to some strange situations)

but as a totally 'empty' block is a valid block for dbv, it doesn't report this as an error.

thank you for your effort,
  Martin

Am 08.09.2009 um 21:37 schrieb Niall Litchfield:

> dbverify?
>
> On Tue, Sep 8, 2009 at 8:03 PM, Martin Berger <martin.a.berger_at_gmail.com
> > wrote:
>
>
> Does someone knows any method to ckeck all blocks of other objects
> (e.g. indexes) by accessing them via their rowid?
>
>
> --
> Niall Litchfield
> Oracle DBA
> http://www.orawin.info



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


  • application/pkcs7-signature attachment: smime.p7s
Received on Tue Sep 08 2009 - 14:53:22 CDT

Original text of this message