Re: Re: Oracle XE Corruption
- From: Andrew Kerber <andrew.kerber@xxxxxxxxx>
- To: daniel.fink@xxxxxxxxxxxxxx
- Date: Wed, 25 Aug 2010 12:15:35 -0500
I agree that it should have been marked as unusable, but it occurs to me
that if it was generating an error message to a dump file, it must not have
been marked as unusable, ie, Oracle must have been trying to use it if it
got the eror message.
On Wed, Aug 25, 2010 at 11:59 AM, Daniel Fink <daniel.fink@xxxxxxxxxxxxxx>wrote:
> I'm a little bothered by the fact that a known corrupt index would be used
> to rebuild itself. I would expect
> 1) The rebuild to fail
> 2) The rebuild to use the table data as the source
>
> One question in my mind is what was the index state? Was it still marked as
> good in the data dictionary despite these errors? Should it not have been
> marked as corrupt/unusable?
>
> ------- Original Message -------
>
> On 8/25/2010 4:52 PM Jared Still wrote:
> On Wed, Aug 25, 2010 at 9:22 AM, Evan Pettrey <jepettrey@xxxxxxxxx> wrote:
>
> The rebuild did work, but the errors still remained and the log continued
>> to grow. After dropping and recreating the indexes everything worked
>> properly.
>>
>>
>> This would seem to validate the statement in the docs that 'REBUILD' does
> read the old index rather than table data.
>
> Someone care to test this?
>
> Win worldwide fame and glory by verifying that the docs are actually
> correct. :)
>
> Jared Still
> Certifiable Oracle DBA and Part Time Perl Evangelist
> Oracle Blog: http://jkstill.blogspot.com
> Home Page: http://jaredstill.com
> --
> //www.freelists.org/webpage/oracle-l
>
>
>
--
Andrew W. Kerber
'If at first you dont succeed, dont take up skydiving.'
Other related posts:
