RE: Read a corrupted block

From: <Joel.Patterson_at_crowley.com>
Date: Fri, 30 Jul 2010 09:13:58 -0400
Message-ID: <0684DA55864E404F8AD2E2EBDFD557DA0545154E_at_JAXMSG01.crowley.com>



I thank everyone for their responses including Frits and Craig who responded off list.  

I have plenty to work with now, and have some exploring to do.  

As a followup: DBMS_REPAIR does not find the block. DUDE cost money and the one block is not worth it in that the data goes away ('essentially') after 90 days - but hangs around for ... some time.  

Best Regards,  

Joel Patterson
Database Administrator
904 727-2546


From: Kurt Van Meerbeeck [mailto:kurtvm_at_telenet.be] Sent: Thursday, July 29, 2010 11:03 AM
To: Patterson, Joel
Cc: oracle-l_at_freelists.org
Subject: Re: Read a corrupted block  

Hi Joel,  

Have you tried just dumping the block using 'alter system dump datafile block' ?

Depending on the corruption it will complete - and starting from 10g the trace files shows enough info to see

what records are involved...

Also metalink note 34371.1 will show you how to find the data involved using rowid range scans.  

kind regards,

Kurt      

  • Originele e-mail ----- Van: "Joel Patterson" <Joel.Patterson_at_crowley.com> Aan: oracle-l_at_freelists.org Verzonden: Donderdag 29 juli 2010 16:46:53 GMT +01:00 Amsterdam / Berlijn / Bern / Rome / Stockholm / Wenen Onderwerp: Read a corrupted block

When all hope fails - block recover, dbms_repair, etc. and before you enable skipping corrupt blocks, is there some utility that can show me what is in the block?  

Something like a hex reader, but since it is only one block, (and we are probably going to be getting rid of that data anyway), the method should be free. (For instance, I think dude (possible spelling) is something that could work, but for one block it would be overkill).  

Even if it is not free, you could let me know. I probably won't get it, but the information would be useful, and the question has been raised.      

--
http://www.freelists.org/webpage/oracle-l
Received on Fri Jul 30 2010 - 08:13:58 CDT

Original text of this message