Re: Oracle Recovery Error

Did you restore the redo logs? I think you did, and this is the problem.

The control file knows that redo file # 147 should still be on-line. Yes, it was copied to the archive directory. But, since redo log # 148 was not yet archived, that means that redo log # 149 (which would overlay redo # 147) had not yet been started. So, rather than ask for archived log # 147, Oracle wants to use redo log # 147.

You almost never need to restore the redo logs. Most of the time it causes this trouble.

gmei_at_my-dejanews.com wrote:

> Hi:
>
> A couple of days ago, I posted a message asking about how to do recovery
> using control file. Thanks for the people who answered my questions. Now I
> have tried to go through a recovery process to see if everything works on our
> system.
>
> 1. First I checked if there is any archived redo log files :
>
> orasrv1> pwd
> /ora02/oradata/PAV1_logarchive
>
> orasrv1> ls -ltr
> -rw-r----- 1 oracle dba 1049600 Apr 26 09:13 pav1archS0000000144.arc
>
> 2. I shutdown the database, did a cold backup of all the files (datafiles,
> control files, logfils, init.ora).
> saved the backup files and “startup open” the db again.
>
> 3. I run a oracle package which run a “update” statatement for 300 times and
> “commit” them so that several redo log files have been generated:
> orasrv1> ls -ltr
> total 4128
> -rw-r----- 1 oracle dba 1049600 Apr 26 09:13 pav1archS0000000144.arc
> -rw-r----- 1 oracle dba 1049600 Apr 26 10:47 pav1archS0000000145.arc
> -rw-r----- 1 oracle dba 1049600 Apr 26 10:47 pav1archS0000000146.arc
> -rw-r----- 1 oracle dba 1049600 Apr 26 10:48 pav1archS0000000147.arc
>
> 4. I shutdown the db, copy the db control file to my directory:
> /home/oracle/pav1ctl.ora
>
> 5. I replaced all db files with the saved backup files from Step 2.
>
> 6. I copied the latest control files to where it should be:
> cp /home/oracle/mycontrol.ora /ora02/oradata/PAV1/pav1ctl.ora
> cp /home/oracle/mycontrol.ora /ora03/oradata/PAV1/pav1ctl.ora
>
> 7. I used svrmgrl, connect internal and startup mount, I got
> SVRMGR> startup mount
> ORACLE instance started.
> Total System Global Area 5074288 bytes
> Fixed Size 60784 bytes
> Variable Size 4456448 bytes
> Database Buffers 409600 bytes
> Redo Buffers 147456 bytes
> Database mounted.
>
> So far so good.
>
> 8. I then issued RECOVER DATABASE and hit return and got:
> SVRMGR> RECOVER DATABASE
> ORA-00279: change 47446 generated at 04/26/99 09:22:20 needed for thread 1
> ORA-00289: suggestion : /ora02/oradata/PAV1_logarchive/pav1archS0000000145.arc
> ORA-00280: change 47446 for thread 1 is in sequence #145
> Specify log: {<RET>=suggested | filename | AUTO | CANCEL}
>
> Log applied. ORA-00279: change 47518 generated at 04/26/99 10:47:17 needed
> for thread 1 ORA-00289: suggestion :
> /ora02/oradata/PAV1_logarchive/pav1archS0000000146.arc ORA-00280: change
> 47518 for thread 1 is in sequence #146 ORA-00278: log file
> '/ora02/oradata/PAV1_logarchive/pav1archS0000000145.arc' noy Specify log:
> {<RET>=suggested | filename | AUTO | CANCEL}
>
> ORA-00283: recovery session canceled due to errors
> ORA-00314: log 1 of thread 1, expected sequence# 147 doesn't match 145
> ORA-00312: online log 1 thread 1: '/ora03/oradata/PAV1/pav1log1.ora'
> ORA-00314: log 1 of thread 1, expected sequence# 147 doesn't match 145
> ORA-00312: online log 1 thread 1: '/ora02/oradata/PAV1/pav1log1.ora'
>
> I also tried a couple of other ways such as
> RECOVER DATABASE UNTIL TIME '1999-04-26:10:45:00' ;
> And got the same error.
>
> So my question is:
>
> Why did I get this error? Did some of my step(s) are incorrect?
>
> I think I am not missing any archived redo log files. Redo log files seems to
> be opend OK. Looking at Oracle doc does not provide any clue to me.
>
> If you reply, please send a copy to zlmei_at_hotmail.com
>
> Thanks in advance.
>
> Guang
>
> -----------== Posted via Deja News, The Discussion Network ==----------
> http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
Received on Mon Apr 26 1999 - 20:30:28 CDT