How can I determine what is the minimum SCN number I need to restore up to.

Question

Say if I have a full database backup, I know I have file inconsistency, but I want to know what is the minimum time or SCN number a need to roll forward to in order to be able to open the database?
For example: I do a database restore.
restore database ;
RMAN> sql 'alter database open read only';
sql statement: alter database open read only
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-03009: failure of sql command on default channel at 03/16/2009 15:00:04
RMAN-11003: failure during parse/execution of SQL statement: alter database open read only
ORA-16004: backup database requires recovery
ORA-01194: file 1 needs more recovery to be consistent
ORA-01110: data file 1: '/u01/oradata/p1/system01.dbf'
I need need to apply archive log files. All references I find for ORA-00194 state the solution is to "apply more logs until the file is consistent " But "HOW MANY LOGS", or more apporaite up to what time or SCN? How does one determine what TIME or SCN is required to get all file consistent?
I thought this query might provide the answer, but it doesn't
select max(checkpoint_change#)
from v$datafile_header
MAX(CHECKPOINT_CHANGE#)
7985876903
--It applies a bit more redo, but not enough to make my datafiles consistent.
recover database until SCN=7985876903 ;
Starting recover at 03/16/09 15:04:54
using channel ORA_DISK_1
using channel ORA_DISK_2
using channel ORA_DISK_3
using channel ORA_DISK_4
using channel ORA_DISK_5
using channel ORA_DISK_6
using channel ORA_DISK_7
using channel ORA_DISK_8
starting media recovery
channel ORA_DISK_1: starting archive log restore to default destination
channel ORA_DISK_1: restoring archive log
archive log thread=1 sequence=18436
channel ORA_DISK_1: reading from backup piece /temp-oracle/backup/hot/p1/20090315/hourly.arch_P1_47353_681538638_1
channel ORA_DISK_1: restored backup piece 1
piece handle=/temp-oracle/backup/hot/p1/20090315/hourly.arch_P1_47353_681538638_1 tag=TAG20090315T041716
channel ORA_DISK_1: restore complete, elapsed time: 00:02:26
archive log filename=/u01/app/oracle/flash_recovery_area/P1/archivelog/2009_03_16/o1_mf_1_18436_4vxd81yc_.arc thread=1 se quence=18436
Oracle Error:
ORA-01547: warning: RECOVER succeeded but OPEN RESETLOGS would get error below
ORA-01194: file 1 needs more recovery to be consistent
ORA-01110: data file 1: '/u01/oradata/p1/system01.dbf'
I've discover I need to apply archive logs until this query reports all datafiles as FUZZY=NO , but this only works by guessing at some time periord to roll forward to, then checking the FUZZY column, and try again. Is there a way to know, I have to roll forward to a specific SNC in order for all my datafiles to be consistent?
select file#
     , status
     , checkpoint_change#
     , checkpoint_time
     , FUZZY
     , RECOVER
,LAST_DEALLOC_SCN
from v$datafile_header
order by checkpoint_time
Thanks,
Jason

Answer 1

The minimum point in time is the time when the last backup piece for datafiles in that backup was completed.
Your alert.log should show the redo log sequence number at that time.
You can query V$ARCHIVED_LOG and get the FIRST_CHANGE# of the first archivedlog generated after that backup piece completed.
A
LIST BACKUP;in RMAN should also show you the SCNs at the time of the backups.
You can also query SCN_TO_TIMESTAMP -- eg
select timestamp_to_scn(to_timestamp('15-MAR-09 09:24:01','DD-MON-RR HH24:MI:SS')) from dual;will return an approximation of the SCN.
Hemant K Chitale
http://hemantoracledba.blogspot.com
Edited by: Hemant K Chitale on Mar 17, 2009 9:41 AM
added the LIST BACKUP command from RMAN.