Re: English semantics help needed

will it ? They talk about redo/archive logs but not about data blocks.

Please consider the environment before printing this e-mail

                                                                                                                                                      
  From:       saurabh manroy <saurabhmanroy_at_gmail.com>                                                                                                
                                                                                                                                                      
  To:         Laimutis.Nedzinskas_at_seb.lt, oracle-l <oracle-l_at_freelists.org>                                                                           
                                                                                                                                                      
  Date:       2011.11.29 13:22                                                                                                                        
                                                                                                                                                      
  Subject:    Re: English semantics help needed                                                                                                       
                                                                                                                                                      

>> the patch does not fix the corruption. It only makes sure the
redo/archive logs are cleaned from this corruption so that backups/standby work.

No,
From the note 7662491.8:
"The fix prevents this corruption to be introduced in the redo and will not fix any existent corruption that is already introduced in the redo/archived logs. Meaning that if the patch is installed and a recovery is attempted again using the same archived logs, the error will be still produced."

>> select from table would still give an error, wouldn't it ?

Yes,Patch wouldn't fix anything that is already corrupt. Note suggests to use dbms_repair to skip corrupt blocks.
>> There is no fix for the root cause ?

Patch would fix the rootcause but that would be of help only in future.

Regards,
Saurabh Manroy

On Tue, Nov 29, 2011 at 12:02 PM, <Laimutis.Nedzinskas_at_seb.lt> wrote:.

      Bug 7662491 - Array Update can corrupt a row. ORA-600
      [kghstack_free1]
      ORA-600 [kddummy_blkchk][6110/6129] [ID 861965.1]

      Hi listers

      I have problem reading the metalink Bug 7662491 and especially patch
      7662491. Btw, we had a similar issue meaning this is not a
      theoretical
      interest of mine.

      Do I interpret the note correctly:

      the patch does not fix the corruption. It only makes sure the
      redo/archive
      logs are cleaned from this corruption so that backups/standby work.
      select from table would still give an error, wouldn't it ?
      There is no fix for the root cause ?


      Brgds, Laimis N



      ---------------------------------------------------------------------------------


      Please consider the environment before printing this e-mail

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

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