Re: block corruption in system tablespace

From: Peter Karl Lichtenwagner <peter_at_monochrom.at>
Date: Tue, 14 Nov 2000 21:23:45 GMT
Message-ID: <B3iQ5.140291$Le2.7876389@news.chello.at>

thank you very much.

however, i don't think the 3gl-trick will work, as the range of the corrupted blocks is somewhere well above 6900 ....

anyway, luckily they do some kind of data exchange with a central database server every night, so there so a chance to reconstruct most of the data ....

thanks again,
peter

Jonathan Lewis <jonathan_at_jlcomp.demon.co.uk> wrote in message news:974233575.16676.0.nnrp-14.9e984b29_at_news.demon.co.uk...
>
> ORA-00704 is an error in bootstrap, which probably
> means that the corrupt blocks are in the CACHE
> segment. There is no LEGAL way around this
> problem.
>
> HOWEVER - you may get lucky.
> Create a whole new database with the same block
> size, and use a 3GL program to read the oracle
> blocks from the new database and write them to
> the same location in the old database. It MIGHT
> be sufficient to allow you to start the database.
>
> On the other hand, if the corrupt blocks fall
> outside the CACHE segment, then you may
> not be able to fix it at all.
>
> --
>
> Jonathan Lewis
> Yet another Oracle-related web site: http://www.jlcomp.demon.co.uk
>
> Practical Oracle 8i: Building Efficient Databases
>
Received on Tue Nov 14 2000 - 15:23:45 CST