Skipping an ArchiveLog in a RESTORE / RECOVER or Standby

I am *surprised* at the number of "so-called" DBAs that keep asking "can I skip an archivelog" during a database recovery or when applying to a standby. They just don't apply their mind to considering what those archivelogs might contain and the concept of time.
If I skip the archivelog that contained the fact that I woke up from sleep, how can I apply the archivelog that contains the fact that I went to work ?
If I skip the archivelog that INSERTed employee 1001 in the EMPLOYEES table, how can I apply the archivelog that contains the subsequent UPDATE to his salary ?
Hemant K Chitale

+1 to all what you said Hemant.
Aman....

Similar Messages

Restore/recover from cold user-managed backup.

i shutdown db,(db is in no-archivelog mode), copy the dir having *.ctl, *.log & *.dbf files, what would be steps to recover in case,
my idea is restore all files in original dir and alter database open resetlogs should do the work,
if that did'n work then mount db and try
recover database using backup controlfile until cancel;
and give path one of online logfile and hit enter; & open db with resetlogs.make me right, if i am not......or any quick way to restore/recover tip?

DBA2008 wrote:
i shutdown db,(db is in no-archivelog mode), copy the dir having *.ctl, *.log & *.dbf files, what would be steps to recover in case,
my idea is restore all files in original dir and alter database open resetlogs should do the work,
if that did'n work then mount db and try
recover database using backup controlfile until cancel;
and give path one of online logfile and hit enter; & open db with resetlogs.make me right, if i am not......or any quick way to restore/recover tip?Why resetlogs ? All what you need is a pfile, if its not there already and start the db. That's all!
Aman....

RAC RMAN Restore/Recover

Hi folks,
I have some doubts about the PATH to disaster recovey with RAC and RMAN.
In this case I have just the RAC without data guard/streams, just only a single rac with two nodes.
The hypothetical scenario is:
One database using ASM (database name orcl) , two inatances in two separated servers (nodes serv1 and serv2) and two instances ( orcl1 and orcl1).
In this scenary I lost everything, and just keep the backups of: spfile, controlfile, database and archives.
So the steps to recovery is (using serv1):
1- Create the cluster (install clusteware, create CRS/Vdisk, create ASM instance)
2 - Just install the oracle files in all nodes serv1 and serv2, but don't create the instance orcl1 and orcl2.
3 - restore/recover spfile - rman
4 - restore/recover controlfile - rman
5 - restore/recover database with until clause - rman
After these steps the instance orcl1 will be restored by RMAN in the serv1 and the database orcl was created in the ASM.
But, please correct if I'm wrong, the instance and database will not be registered in CRS (RMAN not register the instance).
So I need to registry the instance and database with srvctl, doing:
1 - create PFILE in serv1 for orcl1 (that point to SPFILE in ASM) and ORAPWD file.
2 - add the database orcl to CRS
3 - add the instance orcl1 to CRS
4 - add the dependency about ASM and orcl1 from the node serv1 (using srvclt)
But after this I have the instance orcl2 in serv2 to create.
Just the instance, because the binaries already installed and the database was already created.
To complete I need to do the steps:
1 - Copy/Create PFILE for instance orcl2 in serv2
2 - add the instance orcl2 to CRS (using srvctl add instance)
3 - add the dependency about ASM instance (that exists in server2) and Instance orcl2 from the node serv2 (using srvctl)
4 - create PFILE and ORAPWD file.
The question is, RMAN will registry the instancie and database in CRS?
Another question, the RMAN will create PFILE and registry all instances of the original cluster?
I believe that not for two questions, correct?
Exists another path to restore/recover the RAC in this scenary?
The version in this case is 10Gr2 and 11Gr2.
Thanks.

>
The hypothetical scenario is:
One database using ASM (database name orcl) , two inatances in two separated servers (nodes serv1 and serv2) and two instances ( orcl1 and orcl1).
In this scenary I lost everything, and just keep the backups of: spfile, controlfile, database and archives.
So the steps to recovery is (using serv1):
1- Create the cluster (install clusteware, create CRS/Vdisk, create ASM instance)
2 - Just install the oracle files in all nodes serv1 and serv2, but don't create the instance orcl1 and orcl2.
3 - restore/recover spfile - rman
4 - restore/recover controlfile - rman
5 - restore/recover database with until clause - rman
After these steps the instance orcl1 will be restored by RMAN in the serv1 and the database orcl was created in the ASM.
But, please correct if I'm wrong, the instance and database will not be registered in CRS (RMAN not register the instance).
So I need to registry the instance and database with srvctl, doing:
1 - create PFILE in serv1 for orcl1 (that point to SPFILE in ASM) and ORAPWD file.
2 - add the database orcl to CRS
3 - add the instance orcl1 to CRS
4 - add the dependency about ASM and orcl1 from the node serv1 (using srvclt)
But after this I have the instance orcl2 in serv2 to create.
Just the instance, because the binaries already installed and the database was already created.
To complete I need to do the steps:
1 - Copy/Create PFILE for instance orcl2 in serv2
2 - add the instance orcl2 to CRS (using srvctl add instance)
3 - add the dependency about ASM instance (that exists in server2) and Instance orcl2 from the node serv2 (using srvctl)
4 - create PFILE and ORAPWD file.
The question is, RMAN will registry the instancie and database in CRS?NO.. You will need the SRVCTL for registry services.
Another question, the RMAN will create PFILE and registry all instances of the original cluster?No... RMAN perform restore SPFILE only.
I believe that not for two questions, correct?
Correct.
Exists another path to restore/recover the RAC in this scenary?
The version in this case is 10Gr2 and 11Gr2.You can skip a few steps above by doing this:
After installing Oracle Clusterware / RAC
You will configure the network via NETCA.
You will create and configure ASM with DBCA.
To skip all the setup steps the Database (orcl) in Oracle Clusterware, you will create a database via DBCA with the same DB_NAME (orcl) of the database that will be restored.
The DBCA will configure the database with all the dependent services of the Clusterware including TNSNAMES.
Finished creating the database, all services should be configured, you will connect to ASM via asmcmd and drop (spfile, controlfile and datafiles), now just perform the restore (SPFILE,CONTROLFILE and DATAFILES) via rman.

Backing up and Restoring/Recovering an Oracle 9i Database

I am fairly new at this and have been tasked with managing off site backups for customer data on a 9i database.
I am carrying out a user managed backup of the following data files to the offsite data centre.
User File
USERS01.DBF
Control File
CONTROL01.CTL
Redo Log Files
REDO01.LOG
REDO02.LOG
REDO03.LOG
Index File
INDX01.DBF
SPFile
SPFILEBCS.ORA
To test the success of the back up I have created a fresh install of 9i on a new machine with the file structure replicating that of the original.
My questions are;
1) Are these all the files required to be backed up in order to carry out a restore?
2) What are the steps that I need to carry out to restore/recover the database using the above files?
Any help or ‘user friendly’ resources for beginners would be greatly appreciated
Regards
Toby

1. If you are performing a cold backup you must backup all files listed by this query:
SELECT NAME FROM V$CONTROLFILE
UNION
SELECT NAME FROM V$DATAFILE
UNION
SELECT MEMBER FROM V$LOGFILE2. In order for you to restore, you perform a software-only install on the backup host.
3. Configure connectivity (listener.ora, sqlnet.ora, tnsnames.ora)
4. Create the service at the remote host using the oradim command.
5. Restore the above mentioned database file set to the same path they have in the source host.
6. Copy spfileINSTANCE_NAME.ora file to the backup host.
7. Start services
8. Start your database instance
9. Add temporary datafiles to your restored database:
ALTER TABLESPACE tempTS ADD TEMPFILE 'path/name';
10. Test it.

HELP! How can i restore/recover files from Bootcamp with Boot Failure - Missing operating system error?

Hi,
I am having a few problems with bootcamp on my imac. When i tried to move the bootcamp partition on windows 7 into free space, it crashed and turned off. Ever since i have had the error 'Missing operating system' and it fails to boot. I have some very important files on the bootcamp partition. Is it possible to recover these files or even to restore the bootcamp partition to it's original state? I do not have a backup of my bootcamp partition. Another problem i have is that ever since i tried to expand my macintosh HD partition into free space, when i hold alt while booting up (to choose which partition to boot like i normally do), the bootcamp partition has stopped appearing. This means that i cannot even boot the windows 7 bootcamp partition anymore!
So basically here are my 2 questions:
How can i restore/recover files from Bootcamp with Boot Failure - Missing operating system error?
How can i make the Bootcamp partition appear in Startup Disk and when i hold the option key while starting up my computer to allow me to boot windows 7?
Any solutions are very much apreciated!
Thanks!
I know that bootcamp (disk0s4) hasn't been fully deleted since it appears in disk utility (i cant do anything to it though)
And here is the error i see when i try to launch bootcamp (back when i was actually able to boot up bootcamp to the error screen)

WinClone 3 is OS X and saves Windows image it makes for restore - that should work but you will have to try and you would need to make a new image unless it also works with a native Windows system restore image. It is now supported and has come a long way.
http://www.twocanoes.com/
Paragon Clone OS works and does disk-to-disk clone just like CCC you end up with two bootable drives. But does not work with your setup. It would let you clone and move your Windows install to an SSD or another disk drive though and be bootable.
During its clone process it checks for errors which is very helpful and lets you know - something CCC and others should adopt more of.
http://www.paragon-software.com/downloads/demo.html
I wish for our/my sake you had re-read and rewritten the long 'story' and broken it into a brief list of facts we needed.
OS X
Windows
Backup (though external is much safer) and you want bootable OS X clones as well as TimeMachine
https://support.apple.com/kb/HT1427
https://support.apple.com/kb/HT1553
There are a number of things to do like chkdsk and others as well as Windows DVD to do automatic system repairs and find out why.
AppleHFS - the abilty to mount and read HFS volumes can be notorious.
I would rearrange and redo your storage setup and how you use the 4-5 internal hard drive bays.

HT201328 I have paid Bell to unlock my old iphone4 and they have said it is now unlocked. Can I skip the back up and restore process and just buy another carriers sim card from my daugther Kodoo plan, install it and active the iphone4 on itunes account in

I have paid Bell to unlock my old iphone4 and they have said it is now unlocked. Can I skip the back up and restore process and just buy another carriers sim card from my daugther Kodoo plan, install it and active the iphone4 on itunes account in her name?
thanks
Dave

The process is as follows:
Backup the phone, Erase it, connect to iTunes for the unlock process, then your iProfile can be restored.
You can go ahead and get the SIM, bout you have to Erase the phone and connect to iTunes before it will be unlocked.

Restore / Recover / Overall nice treatment of UNDO

Hi,
I'm updating our 'Backup and Recovery Bible' which was written when UNDO was but a twinkle in the eye of Oracle Developers. I'm a bit unsure of how recovery should be done in the event that the UNDO tablespace or any of its datafiles dies.
I do know that I can create another UNDO tablespace if for example we lose one of the datafiles that isn't being used at the time of the failure. But could someone point me to a good link or comment on what should be done if we have a problem with UNDO in terms of restore / recover / RMAN usage etc.
We're using both 9i and 10g on Solaris. For the moment no interface to a media manager.
Cheers.

Confusion sets in.
I don't understand why (after renaming an undo datafile to OFF_undo) I'm not getting anything in the v$recover_file view.
undo datafile name changed
SQL> create table lambda (col1 varchar2(10)) tablespace users;
create table lambda (col1 varchar2(10)) tablespace users
ERROR at line 1:
ORA-00604: error occurred at recursive SQL level 1
ORA-01116: error in opening database file 2
ORA-01110: data file 2: '/database/data01/z111ot11/undo_01.dbf'
ORA-27041: unable to open file
SVR4 Error: 2: No such file or directory
Additional information: 3
SQL> SELECT FILE#, ERROR, ONLINE_STATUS, CHANGE#, TIME
FROM V$RECOVER_FILE; 2
no rows selected
SQL> COL DF# FORMAT 999
SQL> COL DF_NAME FORMAT A35
SQL> COL TBSP_NAME FORMAT A7
SQL> COL STATUS FORMAT A7
SQL> COL ERROR FORMAT A10
SQL> COL CHANGE# FORMAT 99999999
SQL> SELECT r.FILE# AS df#, d.NAME AS df_name, t.NAME AS tbsp_name,
2 d.STATUS, r.ERROR, r.CHANGE#, r.TIME
3 FROM V$RECOVER_FILE r, V$DATAFILE d, V$TABLESPACE t
4 WHERE t.TS# = d.TS#
5 AND d.FILE# = r.FILE#
6 ;
no rows selected
Nothing in the alert.log either. If the UNDO is that important - why isn't the database screaming at me?

Backup & restore/recover Oracle BASE

Before apply upgrade patchset & CUPs i want to take backup of Oracle BASE, any body can send me command/method to do that , and for restore/recover steps also....OS is Solaris8 .....920x are DBs

Is this right?No
cd /product/DBHome
tar cf - * | gzip > /product/archive/DBHome.tar.gzgtar -zcvf /product/archive/DBHome.tar.gz *
>
to restore
Next, delete the Oracle Home:
cd /product
rm -rf /product/DBHome
Restore the backuped Oracle HOME
mkdir -p /product/DBHome
gunzip < c| tar xf -gtar -zxvf /product/archive/DBHome.tar.gz

Computer crashed; new C drive. How do I restore/recover my bookmarks on Sync? Would rather not start from square one.

Computer crashed; new C drive installed. How do I restore/recover my bookmarks on Sync? Would rather not start from square one.

Without the Sync Key, even if you could connect and download your data from the Sync server it would be useless without your Sync Key. The Sync Key is used to encrypt & decrypt the data going to & coming from the Sync server.

RAC Restore/Recover

Hi folks,
I have some doubt about the PATH to disaster recovey with RAC and RMAN.
In this case I have just the RAC without data guard/streams, just only a single rac with two nodes.
The hypothetical scenario is:
One database using ASM (database name orcl) , two inatances in two separated servers (nodes serv1 and serv2) and two instances ( orcl1 and orcl1).
In this scenary I lost everything, and just keep the backups of: spfile, controlfile, database and archives.
So the steps to recovery is (using serv1):
1- Create the cluster (install clusteware, create CRS/Vdisk, create ASM instance)
2 - Just install the oracle files in all nodes serv1 and serv2, but don't create the instance orcl1 and orcl2.
3 - restore/recover spfile - rman
4 - restore/recover controlfile - rman
5 - restore/recover database with until clause - rman
After these steps the instance orcl1 will be restored by RMAN in the serv1 and the database orcl was created in the ASM.
But, please correct if I'm wrong, the instance and database will not be registered in CRS (RMAN not register the instance).
So I need to registry the instance and database with srvctl, doing:
1 - create PFILE in serv1 for orcl1 (that point to SPFILE in ASM) and ORAPWD file.
2 - add the database orcl to CRS
3 - add the instance orcl1 to CRS
4 - add the dependency about ASM and orcl1 from the node serv1 (using srvclt)
But after this I have the instance orcl2 in serv2 to create.
Just the instance, because the binaries already installed and the database was already created.
To complete I need to do the steps:
1 - Copy/Create PFILE for instance orcl2 in serv2
2 - add the instance orcl2 to CRS (using srvctl add instance)
3 - add the dependency about ASM instance (that exists in server2) and Instance orcl2 from the node serv2 (using srvctl)
4 - create PFILE and ORAPWD file.
The question is, RMAN will registry the instancie and database in CRS?
Another question, the RMAN will create PFILE and registry all instances of the original cluster?
I believe that not for two questions, correct?
Exists another path to restore/recover the RAC in this scenary?
The version in this case is 10Gr2 and 11Gr2.
Thanks.

Folks,
Moved to RAC forum thread.

RMAN Restore/Recover

Is it possible to restore an RMAN full database backup to a different machine/db?
I have a Prod DB (800gb) running 10.2.0.4 and am initiating RMAN backup strategy. I am performing full backups to tape and need to prove that the backups are viable for restoration. The problem that I have is that I cannot do any testing on the Prod db. My original plan was to create a new tablespace and datafile - get the rman backup, delete the datafile, and restore from the backup. I am not allowed to do this simple test so it was susggested that we use RMAN to recover from the Prod backup set to a different database on another server. Is this even possible? I have proved functionaly that RMAN can restore in a test environment but management wants to prove we can restore/recover production data wo affecting the primary production db.
Thanks!!!!

Hi.
if you have metalink access take a look to Doc.
388431.1 ( Creating Duplicate Database on a New Host. ) | RMAN
*T

Problem in recover physical standby database(Data Guard) by rman

Hello to all
I have created a physical standby database ,I want make backup of it by rman and when I lose it's datafile I can restore it ,making backup and restore is fine but in recovery I encounter some problem
scenarios is follow
1- In rman I create a backup of standby database by this command:
backup database plus archivelog delete all input;
2- I run this comman in rman for recover standby database
run{
2> set until scn 1392701;
3> restore database;
4> recover database;
5> }
(1392701 is extracted from this query "SELECT MAX(NEXT_CHANGE#)+1 UNTIL_SCN FROM V$LOG_HISTORY LH,
V$DATABASE DB WHERE LH.RESETLOGS_CHANGE#=DB.RESETLOGS_CHANGE# AND LH.RESETLOGS_TIME =
DB.RESETLOGS_TIME;" "http://download.oracle.com/docs/cd/B19306_01/server.102/b14239/rman.htm")
but RMAN result is like this:
executing command: SET until clause
Starting restore at 13-DEC-08
using channel ORA_DISK_1
channel ORA_DISK_1: starting datafile backupset restore
channel ORA_DISK_1: specifying datafile(s) to restore from
backup set
restoring datafile 00001 to /u01/app/oracle/oradata/sari/system01.dbf
restoring datafile 00002 to /u01/app/oracle/oradata/sari/undotbs01.dbf
restoring datafile 00003 to /u01/app/oracle/oradata/sari/sysaux01.dbf
restoring datafile 00004 to /u01/app/oracle/oradata/sari/users01.dbf
restoring datafile 00005 to /u01/app/oracle/oradata/sari/example01.dbf
restoring datafile 00006 to /u01/app/oracle/oradata/sari/users02.dbf
channel ORA_DISK_1: reading from backup piece /home/oracle/backup/0ek24dt4_1_1
channel ORA_DISK_1: restored backup piece 1
piece handle=/home/oracle/backup/0ek24dt4_1_1
tag=TAG20081213T042506
channel ORA_DISK_1: restore complete, elapsed time: 00:01:07
Finished restore at 13-DEC-08
Starting recover at 13-DEC-08
using channel ORA_DISK_1
starting media recovery
archive log thread 1 sequence 116 is already on disk as file /u01/app/oracle/oradata/archive/1_116_666786084.arc
archive log thread 1 sequence 117 is already on disk as file /u01/app/oracle/oradata/archive/1_117_666786084.arc
archive log filename=/u01/app/oracle/oradata/archive/1_116_666786084.arc thread=1 sequence=116
archive log filename=/u01/app/oracle/oradata/archive/1_117_666786084.arc thread=1 sequence=117
unable to find archive log
archive log thread=1 sequence=118
RMAN-03002: failure of recover command at 12/13/2008 05:14:13
RMAN-06054: media recovery requesting unknown log: thread 1
seq 118 lowscn 1392700
3- then I decline 1392701 to 1392700 and i run this command
run{
2> set until scn 1392700;
3> restore database ;
4> recover database;
5> }
executing command: SET until clause
Starting restore at 13-DEC-08
using channel ORA_DISK_1
channel ORA_DISK_1: starting datafile backupset restore
channel ORA_DISK_1: specifying datafile(s) to restore from
backup set
restoring datafile 00001 to /u01/app/oracle/oradata/sari/system01.dbf
restoring datafile 00002 to /u01/app/oracle/oradata/sari/undotbs01.dbf
restoring datafile 00003 to /u01/app/oracle/oradata/sari/sysaux01.dbf
restoring datafile 00004 to /u01/app/oracle/oradata/sari/users01.dbf
restoring datafile 00005 to /u01/app/oracle/oradata/sari/example01.dbf
restoring datafile 00006 to /u01/app/oracle/oradata/sari/users02.dbf
channel ORA_DISK_1: reading from backup piece /home/oracle/backup/0ek24dt4_1_1
channel ORA_DISK_1: restored backup piece 1
piece handle=/home/oracle/backup/0ek24dt4_1_1 tag=TAG20081213T042506
channel ORA_DISK_1: restore complete, elapsed time: 00:01:08
Finished restore at 13-DEC-08
Starting recover at 13-DEC-08
using channel ORA_DISK_1
starting media recovery
archive log thread 1 sequence 116 is already on disk as
file /u01/app/oracle/oradata/archive/1_116_666786084.arc
archive log thread 1 sequence 117 is already on disk as
file /u01/app/oracle/oradata/archive/1_117_666786084.arc
archive log filename=/u01/app/oracle/oradata/archive/1_116_666786084.arc thread=1
sequence=116archive log
filename=/u01/app/oracle/oradata/archive/1_117_666786084.arc
thread=1 sequence=117Oracle Error:
ORA-01547: warning: RECOVER succeeded but OPEN RESETLOGS
would get error below
ORA-01152: file 1 was not restored from a sufficiently old backup
ORA-01110: data file 1: '/u01/app/oracle/oradata/sari/system01.dbf'
media recovery complete, elapsed time: 00:00:10
Finished recover at 13-DEC-08
4- if I run
run{
restore database;
recover database;
I will recieve that error of step 2 (RMAN-06054: media recovery requesting unknown log: thread 1
seq 118 lowscn 1392700)
5- if I just restore the database and I don't perform recovery by rman and I restart redo apply all thing seem fine
but in opening database I'll recieve ORA-01152: file 1 was not restored from a sufficiently old backup
ORA-01110: data file 1: '/u01/app/oracle/oradata/sari/system01.dbf' error)
do you know what is problem
thanks
Edited by: ARKH on Dec 12, 2008 11:06 PM

hi
I myself have found the solution , when I recover the standby database
it do recovery but at the end of recovery it raise the error(RMAN-06054: media recovery requesting unknown log: thread 1
seq 118 lowscn 1392700) but if I begain redo apply before open the database
and I wait till all redo apply process start and communication between the
standby database and the primary database start, then I can
open the standby database and no error will raise
but if befor restarting redo apply I open the database I'll recieve the
ORA-01152: file 1 was not restored from a sufficiently old backup
ORA-01110: data file 1: '/u01/app/oracle/oradata/sari/system01.dbf' error
thanks

Recover the standby database

Hi, I have both primary and standby database in ASM and I need to recover the standby database as it's falling behind for too long. The steps I did are:
1. backup full database on primary
2. create a standby controlfile and copy to /tmp on the standby
3. on the standby, RMAN>restore standby controlfile from '/tmp/std.ctl';
4. RMAN>restore database;
5. SQL>recover managed standby database disconnect;
However, the recovery always shuts down and I get the error:
ORA-01110: data file 1: '+W08_DATA_DG/w08p/datafile/system.295.662256331'
What's wrong with my steps?
Your suggestion is greatly appreciated!

the directory structures are the same. The standby was working before. It's just falling behind so I want to restore it from the new backup. Here is the init.ora
*.audit_file_dest='/opt/app/oracle/admin/W08P/adump'
*.background_dump_dest='/opt/app/oracle/admin/W08P/bdump'
*.cluster_database_instances=2
*.cluster_database=false
*.compatible='10.2.0.3.0'
*.control_files='+W08_DATA_DG/w08p/control01.ctl','+W08_RECOVERY_DG/w08p/control02.ctl','+W08_DATA_DG/w08p/control03.ctl'
*.core_dump_dest='/opt/app/oracle/admin/W08P/cdump'
*.db_block_size=8192
*.db_create_file_dest='+W08_DATA_DG'
*.db_domain=''
*.db_file_multiblock_read_count=16
*.db_name='W08P'
*.dispatchers='(PROTOCOL=TCP) (SERVICE=W08PXDB)'
*.event='10298 trace name context forever, level 32'
*.job_queue_processes=10
*.log_archive_dest_1='LOCATION=+W08_RECOVERY_DG/W08P/'
*.log_archive_format='%t_%s_%r.arc'
*.log_archive_max_processes=6
*.open_cursors=1000
*.optimizer_index_caching=70
*.optimizer_index_cost_adj=30
*.pga_aggregate_target=1459617792
*.processes=300
*.remote_login_passwordfile='exclusive'
*.sessions=1000
*.sga_target=1610612736
*.undo_management='AUTO'
*.undo_retention=26000
W08P.undo_tablespace='UNDOTBS1'
*.user_dump_dest='/opt/app/oracle/admin/W08P/udump'
standby_archive_dest='LOCATION=+W08_RECOVERY_DG/W08P'
CONTROL_FILE_RECORD_KEEP_TIME=0
standby_file_management=auto
fal_server=w08p
fal_client=w08p_stby
control_file_record_keep_time=90
~

Restore production from standby database

Hi,
I would like to make a standby database for my production database.
But, I have a basic question to restore my production database from standby database. There is a method to make it ?
If I lost all files from production database, how to restore from my standby datase (which become in this case my new production database) ?
thanks in advance for your help.
gnom92.

* I make a copy datafiles from my new production database (old standby database) on old production (where crash)
* With my old production database, I make a standby database on new production database (old standby)
* I can close the new production database and open the new standby database in read / write mode, and this is my production database.
* After what I recreate my standby database (old standby,newproduction), or can I return in recover standby mode ?
thank you very much for your patience.
gnom92.

Trying to recover an standby database

- Oracle 10.2.0.4 Standard Edition
- O/S: Windows Server 2003
- Phisically Standby
We have primary and a standby database. My standby database was synchronized through copying archive redo logs from Primary to Standby. They were copied every 15 min to standby server and then they were recovered from there. Because a lot of archive logs were missed for ever, now my standby database is not synchronized with my primary.
In order to recover my Standby, one week ago I followed these instructions:
Step 1: On the standby database check the current scn.
STDBY> select current_scn from v$database;
CURRENT_SCN
2527075174
Step 2: On the primary database create the needed incremental backup from the above SCN
C:\Documents and Settings\frego>rman target /
run {
allocate channel c1 type disk;
BACKUP INCREMENTAL FROM SCN 2527075174 DATABASE
format '\\srvbackup\archiveslogs-35\arch\incr_bkp_%U';
Step 3: Cancel managed recovery at the standby database
STDBY>alter database recover managed standby database cancel;
Media recovery complete.
Copied the backup files to standby server to "D:\arch\" folder.
Step 4: Catalog the Incremental Backup Files at the Standby Database
D:\arch> rman target /
RMAN> CATALOG START WITH 'D:\arch\';
Step 5: Apply the Incremental Backup to the Standby Database
RMAN> RECOVER DATABASE NOREDO;
Step 6: Put the standby database back to managed recovery mode.
STDBY>> recover managed standby database disconnect;
Media recovery complete.
Step 7: On the primary create new standby controlfile
PRIM>alter database create standby controlfile as 'C:\incr_bkp\standby01.ctl';
System altered.
Step 8: At Standby .. Replace standby controlfile at all location as shown by controle_files parameter.
Copy the standby control file to the standby site. Shutdown the standby database and replace the stanby controlfiles and restart the >standby database in managed recovery mode...
Problem was that there were new datafiles created in primary that didn't exist in standby. I had to created them manually on standby:
ALTER DATABASE CREATE DATAFILE 'D:\ORADATA\PROD\OPEXT\DATAFILES_DATA\SIPESG_BLOB3.ORA' AS 'D:\ORADATA\PROD\OPEXT\DATAFILES_DATA\SIPESG_BLOB3.ORA';
ALTER DATABASE CREATE DATAFILE 'D:\ORADATA\PROD\OPEXT\DATAFILES_DATA\SIPESG_BLOB4.ORA' AS 'D:\ORADATA\PROD\OPEXT\DATAFILES_DATA\SIPESG_BLOB4.ORA';
ALTER DATABASE CREATE DATAFILE 'D:\ORADATA\PROD\OPEXT\DATAFILES_DATA\SIPESG_BKP.ORA' AS 'D:\ORADATA\PROD\OPEXT\DATAFILES_DATA\SIPESG_BKP.ORA';
ALTER DATABASE CREATE DATAFILE 'D:\ORADATA\PROD\OPEXT\DATAFILES_DATA\SIPESG_BLOB5.ORA' AS 'D:\ORADATA\PROD\OPEXT\DATAFILES_DATA\SIPESG_BLOB5.ORA';
ALTER DATABASE CREATE DATAFILE 'D:\ORADATA\PROD\OPEXT\DATAFILES_DATA\SIPESG_BLOB_A.ORA' AS 'D:\ORADATA\PROD\OPEXT\DATAFILES_DATA\SIPESG_BLOB_A.ORA';
ALTER DATABASE CREATE DATAFILE 'D:\ORADATA\PROD\OPEXT\DATAFILES_DATA\SIPESG_BLOB_B.ORA' AS 'D:\ORADATA\PROD\OPEXT\DATAFILES_DATA\SIPESG_BLOB_B.ORA';
ALTER DATABASE CREATE DATAFILE 'D:\ORADATA\PROD\OPEXT\DATAFILES_DATA\SIPESG_BLOB_C.ORA' AS 'D:\ORADATA\PROD\OPEXT\DATAFILES_DATA\SIPESG_BLOB_C.ORA';
ALTER DATABASE CREATE DATAFILE 'D:\ORADATA\PROD\OPEXT\DATAFILES_DATA\SIPESG_BLOB_D.ORA' AS 'D:\ORADATA\PROD\OPEXT\DATAFILES_DATA\SIPESG_BLOB_D.ORA';
ALTER DATABASE CREATE DATAFILE 'D:\ORADATA\PROD\OPEXT\DATAFILES_IDX\PADRONIDX17.ORA' AS 'D:\ORADATA\PROD\OPEXT\DATAFILES_IDX\PADRONIDX17.ORA';
Also, I tried this commands:
On Standby:
RMAN> SHUTDOWN;
RMAN> STARTUP NOMOUNT;
RMAN> RESTORE STANDBY CONTROLFILE FROM 'D:\arch\standby01.ctl';
RMAN> SHUTDOWN;
RMAN> STARTUP MOUNT;
STDB> ALTER DATABASE CLEAR LOGFILE GROUP 1;
STDB> ALTER DATABASE CLEAR LOGFILE GROUP 2;
STDB> ALTER DATABASE CLEAR LOGFILE GROUP 3;
STDB> ALTER DATABASE RECOVER MANAGED STANDBY DATABASE DISCONNECT;
Database altered.
SQL> alter database open read only;
alter database open read only
ERROR at line 1:
ORA-01154: database busy. Open, close, mount, and dismount not allowed now
SQL> recover managed standby database cancel;
Media recovery complete.
SQL> alter database open read only;
alter database open read only
ERROR at line 1:
ORA-16004: backup database requires recovery
ORA-01152: file 1 was not restored from a sufficiently old backup
ORA-01110: data file 1: 'D:\ORADATA\PROD\OPEXT\SYSTEM01.DBF'My Questions:
- What is missing in order to open standby database?
- Can I again generate rman incremental backups on Primary (from SCN 2527075174), copy them to Standby and start recovery again? Would it generate duplicate records on tables?
- Should I generate new rman incremental backup pieces on Primary (from last SCN of Standby), copy them to standby and apply there?
Remember that my Primary is growing every day and those rman backup pieces and standby01.ctl, that are copied on my standby server, are from one week ago.
Thanks
Edited by: user521219 on 22-jun-2012 9:31

user521219 wrote:
Should I restart and mount my standby database again?
If I apply Step 3 now:
STB> alter database recover managed standby database cancel;
ERROR at line 1:
ORA-16136: Managed Standby Recovery not activeRefer the above mentioned link. First restore controlfile & then perform recovery on standby database.
>
user521219
     Newbie
Handle:      user521219
Status Level:      Newbie
Registered:      Jul 20, 2006
Total Posts:      48
Total Questions:      29 (23 unresolved)
>
Consider closing your threads and keep the forum clean.
Edited by: CKPT on Jun 22, 2012 10:43 PM

Skipping an ArchiveLog in a RESTORE / RECOVER or Standby

Similar Messages

Maybe you are looking for