Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: corruption on redo log files

Re: corruption on redo log files

From: Samantha Hall <samanthahall_at_qwest.net>
Date: Thu, 31 May 2001 22:46:24 -0600
Message-ID: <3B171E20.B51EFB7E@qwest.net>

Wayne,

As far as I know, the only way to get your database open is to do a incomplete recovery, and open with a resetlogs.

First you have to ask yourself, how you got corruptions in both members of a logfile group. Normally, if it is a disk corruption problem, and each member of the group are on different disks, you just shut the databse down, and copy the good file over the old one, making sure you copy all the files out of the way first, just in case you get it wrong.

If you know that both are corrupted, which looks the way according to the errors, you need to look for the SCN or time that this log starts, and recover up until that point. You can find this time from your alert.log or from v$log/v$log_history with your database mounted. An open resetlogs will reset the reset the log sequence number to 1 and recreate the logfiles for you. It also does a number of other admin tasks to the controlfile and file headers. Make sure that all your datafiles are ONLINE before issuing this, otherwise they become unavailable needing media recovery to be complete. If you are not running in archivelog mode, these files are lost.

As soon as the database is open with resetlogs, shutdown normal or immediate and take a full cold backup. Backups before this time can no longer be used to recover through the resetlogs operation.

Good luck,
Samantha

Cantonese Boy wrote:

> Hi,
>
> My most current redo log files was corrupted and my
> database doesn't turn on the archive log mode.
> Is it possible to recover it?
>
> I have try to clear the redo log file but it doesn't work.
>
> SVRMGR> alter database open;
> alter database open
> *
> ORA-00355: change numbers out of order
> ORA-00353: log corruption near block 2 change 545248014 time 05/19/01
> 03:50:04
> ORA-00312: online log 5 thread 2: '/dev/rhl_redo0502.lg'
> ORA-00312: online log 5 thread 2: '/dev/rhl_redo0501.lg'
> SVRMGR> alter database clear unarchived logfile group 5;
> alter database clear unarchived logfile group 5
> *
> ORA-00600: internal error code, arguments: [2752], [2], [3], [], [], [],
> [], []
>
> I'm using oracle 7.4.3 on AIX 4.2.1
>
> Thanks for any suggestion
>
> Wayne
Received on Thu May 31 2001 - 23:46:24 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US