Recovery of RMAN backup in third instance
Hi,
is it possible for RMAN to create backups and use these backups to recover the database
in another instance.
For example as a disaster recovery. Is it possible to use the RMAN backups to create a new instance
with the data created by the RMAN backup.
Short answer, perhaps with doc link, is very welcome. Thanks
Short Answer : Yes, Of Course !
Links :
http://download.oracle.com/docs/cd/B19306_01/backup.102/b14191/rcmrecov.htm#i1006453
http://download.oracle.com/docs/cd/B19306_01/backup.102/b14191/rcmrecov.htm#i1007814
Hemant K Chitale
http://hemantoracledba.blogspot.com
Similar Messages
-
Problem in recovery from RMAN backup in case of disk failure
Hi
I have Oracle 9.2.0.1.0 on Windows. Full Hot RMAN backup of database is scheduled on daily basis. I am taking backup on disks. And then i do manually copy them on tapes. If any disk failure occurs than can i recover database completely in this scenario. If tried restoing that backup on different machine as given below
set ORACLE_SID=<name>
rman target sys/<pwd>@ORACLE_SID nocatalog
startup nomount
restore spfile
create pfile from spfile
startup nomount pfile
alter database mount
restore database
Everything worked fine till above steps. Now i will execute last step.
recover database
Here it gives error:-RMAN-06054: media recovery requesting unknown log: thread 1 scn 264750
Still i typed
alter database open resetlogs
Now i can open my database, But is it recovered completely? Or there is loss of some data? Why it is giving this error?
Thanks in advance. Waiting for reply.I can understand the confusion with this whole concept.
What I meant by all archivelogs is all the archivelogs that were backed up with the backup itself. Again, it is essential to ensure that you archive the current redo log AFTER the backup is performed and then back those archivelogs too. Good thing is, RMAN will take care of this for you if you backup the archivelogs with the BACKUP DATABASE command and/or backup the archivelogs AFTER you run the BACKUP DATABASE command.
Once you've mounted the controlfile and CATALOG the archivelogs, you can go into SQL*Plus and issue the following query:
select SEQUENCE#, FIRST_CHANGE#, NEXT_CHANGE# from v$archived_log
order by FIRST_TIME;This will tell you what archivelog is the latest. Logs generated during the backup may also provide the necessary information. Performing Disaster Recovery should help to understand the full process.
The ONLINELOG directory in FRA is only populated if you've configured your redo logs to be stored there. However, you are on 9i...right? If so 9i doesn't utilize a FRA.
HTH
Edited by: ebrian on Oct 3, 2008 1:08 PM -
Undo tablespace recovery from RMAN backup
Hi,
Lets's assume we have RMAN backup and archivelog backups, and now
my undo tablespace's disk crashed ,how can i recover?
Can i recover while database is up?
Thanks,
Kumar.to recover UNDO tablespace from RMAN backup you must shutdown the database and start it in the MOUNT state to recover the missing files as follows
rman target /
RUN
STARTUP MOUNT;
ALLOCATE CHANNEL ch3 TYPE Disk;
RESTORE datafile 'c:\u01\prod3\undo01.dbf';
RECOVER datafile 'c:\u01\prod3\undo01.dbf';
ALTER DATABASE OPEN;
RELEASE CHANNEL ch3;
hope this will help you -
Load rman backup on another instance
Hi All,
I am relatively new to Oracle and now I have to devise a strategy for backups and recovery. My objective is to backup the production database regularly in such a way that one month down the line I am able to load it on another instance. I have been trying to read up on the subject, but looks like it can be loaded only on the same instance.
Any thoughts how can this be achieved.
Thanks
SanjayHi Pierre,
Looks like rman duplicate database command only supports creation of duplicate database on a instance that exists at the time. I am looking for something that can be archived on to disk/tape and when required restored on a database instance. I come from Sybase background, where we just dump (or backup) a database copy and restore it on to another server and database. I have tried dumping schema using both datapump and exp, but both have problems while restoring on another server. Datapump does not support XMLtype data and imp has problem restoring some views. -
Hi ,
i have rman level 0 backup on sunday, and level 1 backup monday to friday.
if i want to recover from these backups until friday , which archivelogs will be required?
To be more specific i require archivelogs of only friday or all archivelogs from sunday through friday?
Thanks
Rajkumaruser615800 wrote:
Hi ,
i have rman level 0 backup on sunday, and level 1 backup monday to friday.
if i want to recover from these backups until friday , which archivelogs will be required?
To be more specific i require archivelogs of only friday or all archivelogs from sunday through friday?
Thanks
RajkumarOnly require fridays archivelogs -
RMAN Backup Issue when Full backup file was missing
Hi,
On : 11.2.0.2 version, Recovery Manager
RMAN Backup issue
Last week we had the NFS Issue where we backup the files onto NFS.
The problem occurred in the below sequence
Saturday the backup was successful
Saturday before deleting the backup files they are copied to tape
Deleting the backup files via OS command
Sunday Full backup started but due the NFS issue backup failed(Killed the instance and restarted)
Did the RMAN Incremental backup which worked on Monday
On Tuesday RMAN started complaining and started doing the Full backup.
no parent backup or copy of datafile 1 found
no parent backup or copy of datafile 3 found
no parent backup or copy of datafile 24 foundSony any idea why the Incremental backup worked on Monday but not on Tuesday
Regards
NMOk here is why I would like exact filenames..
1) to speed the time of recovery. I am much more familiar with traditional recovery methods than RMAN, and while I myself installed and suggested RMAN for backups, our other (11i) database does not use RMAN and so I am more familiar with traditional recovery. So to be able to recover quickly I want to have these files as close as possible to the original names so that I can use my standard script (from a trace file) to recreate the control fiile. If I can get RMAN to spit out the filies with original names but keep RMAN for making the backup, I may ditch the rman bacjup script in chron and write a traditional hot backup script which copies the files the old fashioned way.
2) to reduce cloning time. The RMAN duplicate command is not as good as I thought. I was actually easier for me to restore RMAN backups first to a box that has all the same paths, and THEN shutdown (cold) these files and copy them cold to another server, and restore from there. Agaiin if I can get the original file names, I could choose to do recovery either way (RMAN or traditional) in a real emergency instead of being forced to use rman.
3) to be able to switch to a backup file in the FRA without having to rename it. Seems logical if it is already named correctly.
thanks ! Marvi -
ORACLE8: RECOVERY MANAGER 사용 방법(RMAN)
제품 : ORACLE SERVER
작성날짜 : 1997-11-17
Oracle 8 NEW : Recovery manager 란 ?
====================================
- 소개
Oracle 8 은 새로운 backup와 recovery 방법인 Recovery manager 를 제공한다.
Recovery manager(이하 RMAN) 는 backup,restore,recover process를 관리하는
tool 이다. Oracle Enterprise Manager 를 통해 GUI mode 로 가능하고,
rman 이라는 command line interface 를 함께 제공한다.
특히, OS device control 기능이 보완 되었기 때문에, 현재 backup catalog 를
유지하고, 이 정보를 이용해 OS level 에서 file 을 auto-restore 해 주기
때문에, database recovery 전에 file 을 manual 하게 restore 할 필요가
없어졌다.
Incremental backup 을 제공하고, backup 이나 restore 중에 current block
detection 을 해 주며(v$backup_corruption, v$copy_corruption 또는 alert
log 에 report), Rman 작업시에 parallelization 을 제공하며, open database
backup 동안 특별한 redo 를 발생하지 않는다.
- 구조
Recovery
manager in OEM(oracle
enterprise ---> Database
manager) <--- (catalog)
or RMAN
/| |
| |/
DATABASE ---> OS
(target to backup) <--- Device
- 구성 요소 설명
1) recovery manager
Command 를 해석하고, backup, restore, recover 를 실행하기 위해 target
database 로 command 를 전달한다. recovery catalog 를 update 한다. 실행
과정을 조정, 모니터링 한다.
2) Target database
backup,restore,recovery action 이 수행될 데이타베이스
3) Recovery Catalog
Information 저장장소. 즉, target database 의 물리적 스키마, datafile 과
archivelog 의 backup sets 과 pieces, backup script 등을 포함하고 있음.
rman 작업시 recovery catalog 정보를 이용한다.
4) channel
allocation channel 은 target database 의 backup,restore,recover 에 대한
server process 초기화를 한다. 즉, 이 channel 은 disk 를 포함 기타 OS
device 를 지정하게 되며, 이 갯수에 따라 parallelization 의 degree 가 결정
된다.
- recovery catalog 를 사용할 것인지를 결정하기
catalog 없이 RMAN 을 사용할 때의 단점은 recovery catalog 의 overhead 가
없는 대신, Point-In-Time recovery 를 쉽게 할 수 없다. 또한, control file
손상시에 recovery 할 수 없고, stored script 를 사용할 수 없다.
(datafile 이 20개 이상일 경우에는 recovery catalog 사용을 권장한다.)
- Setup
가. Recovery catalog 생성
1. Backup(target) db 외의 별도의 db 생성 (8.0.3 이상)
2. catalog DB 에 분리된 tablespace 생성
create tablespace rcvcat datafile '/oracle/rcvcat/rcvcat01.dbf'
size 20 M;
3. catalog 의 owner 생성과 권한부여
create user rman identified by rman
temporary tablespace temp
default tablespace rcvcat;
grant dba,connect,resource,recovery_catalog_owner to rman;
4. catalog 저장에 필요한 object 생성
cd $ORACLE_HOME/rdbms/admin
sqlplus rman/rman
@catrman
나. Recovery manager 기동
1. recovery catalog DB 를 사용하는 경우
rman target \"system/manager\" rcvcat \"rman/rman@orabeta1\"
(orabeta1 은 catalog DB 로 접속하는 tnsname 입니다.
즉, sqlplus rman/rman@orabeta1 하면, catalog db 와 접속되어야 합니다.)
2. recovery catalog DB 를 사용하지 않는 경우
rman target system/manager@target nocatalog
위 경우 모두, target database 를 catalog 에 등록해야 한다.
RMAN> register database;
다. Backup 시작
- Backup 은 database 가 open 또는 close(close 후에는 반드시 mount
상태이어야 한다)
- Backup 의 단위
full database
tablespace
datafile
control file
archived logs
- Backup 의 예
--full database
run {
allocate channel dev1 type disk;
allocate channel dev2 type disk;
backup full filesperset 2
(database format '/user7/ora8/backup.%n_%p') ;
release channel dev1;
release channel dev2;
-- tablespace
run {
allocate channel dev_1 type disk ;
backup incremental level 0
filesperset 1
(tablespace system channel dev_1 format '/vol312/net8%d_%u');
release channel dev_1;
-- datafile
run{
allocate channel dev1 type disk;
copy datafile "/vo3/orabeta6/backup/copy_tools.dbf"
to "/vo3/orabeta6/oradata/orabeta6/tools01.dbf"
tag "org_tools";
release channel dev1;
-- archived logs
create script beta_arch_full {
allocate channel dev1 type disk;
allocate channel dev2 type disk;
backup full filesperset 10
(archivelog all delete input
format '/vo3/orabeta6/backup/beta_arch.%s_%p') ;
release channel dev1;
release channel dev2;
-- backup to tape
run {
allocate channel t1 type 'sbt_tape';
backup
format 'dbfull_sunday_s%s_t%t'
tag 'Sunday full backup'
(database);
라. Restore
- Backup 으로 부터 원래 위치로 복원
- Archive log file 은 manual 하게 restore 할 수도 있고, 필요하다면
recovery 시 자동으로 restore 된다.
restore (archivelog all| like <filename | <archivelog range>);
- restore 명령
database : restore (database);
tablespace : restore (tablespace <name>,...,<name>);
datafile : restore (datafile <name>,...,<name>);
control file : restore control file to location;
- recovery 시에 database 는 어떤 상태에 있어야 하는가 ?
1) control file restore - not mount
SVRMGRL> startup nomount
2) whole database, or system tablespace - not open
SVRMGRL> startup mount
3) tablespace or datafile - open but tablespace or datafile offline
SVRMGRL> alter tablespace user_data offline temporary;
SVRMGRL> alter database datafile '/oracle/file/user_data1.dbf' offline;
마, Recovery
- Recoervy 는 file restore 후의 과정이다.
1) Archive log mode : complete recovery, incomplete recovery 가 가능하다
2) No archive log mode : consistent whole database backup 으로 부터
restore 하는 것이 유일한 option 이다.
- whole database recovery 의 예
svrmgrl
connect internal
shutdown abort
startup mount
rman target \"system/manager\" rcvcat \"rman/rman@orabeta1\"
run {
allocate channel dev1 type disk;
allocate channel dev2 type disk;
restore database;
recover database;
release channel dev1;
release channel dev2;
or
run {
allocate channel dev1 type disk;
allocate channel dev2 type disk;
restore tablespace 'SYSTEM';
restore tablespace 'USERS';
recover tablespace 'SYSTEM';
recover tablespace 'USERS';
sql "alter database open ";
release channel dev1;
release channel dev2;
- datafile 하나가 사용할 수 없을 경우의 recovery
가정) full backup set 이 있을 경우
ORA-01157: cannot identify data file 3 - file not found
ORA-01110: data file 3: '/user7/ora8/oradata/ORA8/temp01.dbf'
SVRMGR> startup mount
rman target \"system/manager\" rcvcat \"rman/rman@orabeta1\"
RMAN> run {
allocate channel dev1 type disk;
restore database;
recover database;
release channel dev1;
};Check the TSM logs for errors, the error stack here doesn't show any useful information (only informational messages and an undefined exit code). This is an 8i problem, later releases transfer much better messages from the media management software.
-
Hi All,
I need details about some diagnostic test regarding RMAN backup on my instance.
The following are the current configuration of details:
SQL> select release_name from fnd_product_groups;
RELEASE_NAME
11.5.8
SQL> select * from v$version;
BANNER
Oracle9i Enterprise Edition Release 9.2.0.8.0 - 64bit Production
PL/SQL Release 9.2.0.8.0 - Production
CORE 9.2.0.8.0 Production
TNS for Solaris: Version 9.2.0.8.0 - Production
NLSRTL Version 9.2.0.8.0 - Production
===============================================
Database Size
1.5 TB
Earlier the RMAN backup used to take 20 -23 hours, now for the last 2 months, it has started taking 40 hours.
We want to identify the reason behind the increase in time.
Please let us know methods that we can use to diagnose the root cause.
Thanks
PeeksHi Talip,
Thanks for you suggestions.
I tried the three queries suggested by you, please find below the output:
SQL> select set_stamp from v$backup_async_io where to_date(open_time,'dd.mon.rr')='05.mar.12';
SET_STAMP
777154038
777154038
777154038
777154038
777154038
777154038
777154038
777154038
777154038
777154038
777167658
SET_STAMP
777167658
777167658
777167673
777167673
777167673
16 rows selected.The we checked the effective_bytes_per_second "mb/sec" column for the 777154038 and 777167658 Set_stamps.
SQL> select device_type "Device",
type,
open_time,close_time,
elapsed_time/100 "Ela",
effective_bytes_per_second/1024/1024 "mb/sec",
filename
from v$backup_async_io
where set_stamp = &a
order by close_time desc; 2 3 4 5 6 7 8 9
Enter value for a: 777154038
old 8: where set_stamp = &a
new 8: where set_stamp = 777154038
Device TYPE OPEN_TIME CLOSE_TIME Ela mb/sec FILENAME
SBT_TAPE OUTPUT 05-MAR-12 05-MAR-12 13498 3.03509712 bk_87606_1_777154038
SBT_TAPE OUTPUT 05-MAR-12 05-MAR-12 13475 3.03425884 bk_87605_1_777154038
DISK AGGREGATE 05-MAR-12 05-MAR-12 13476 3.041852
DISK INPUT 05-MAR-12 05-MAR-12 2832 2.90395451 /u1141/oradata/PRD1/rbs03.dbf
DISK AGGREGATE 05-MAR-12 05-MAR-12 13446 3.0486393
DISK INPUT 05-MAR-12 05-MAR-12 2829 2.90703392 /u1141/oradata/PRD1/rbs01.dbf
DISK INPUT 05-MAR-12 05-MAR-12 5332 3.07276821 /u1140/oradata/PRD1/ard01.dbf
DISK INPUT 05-MAR-12 05-MAR-12 5314 3.08317661 /u1140/oradata/PRD1/applsysx01.dbf
DISK INPUT 05-MAR-12 05-MAR-12 5309 3.08608055 /u1140/oradata/PRD1/gld01.dbf
DISK INPUT 05-MAR-12 05-MAR-12 5300 3.09132099 /u1140/oradata/PRD1/applsysd01.dbf
10 rows selected.
SQL> /
Enter value for a: 777167658
old 8: where set_stamp = &a
new 8: where set_stamp = 777167658
Device TYPE OPEN_TIME CLOSE_TIME Ela mb/sec FILENAME
SBT_TAPE OUTPUT 05-MAR-12 06-MAR-12 13323 3.07180786 bk_87607_1_777167658
DISK AGGREGATE 05-MAR-12 06-MAR-12 13289 3.08224869
DISK INPUT 06-MAR-12 06-MAR-12 2652 3.08898926 /u1141/oradata/PRD1/rbs02.dbf
DISK INPUT 06-MAR-12 06-MAR-12 5330 3.0739212 /u1140/oradata/PRD1/invd01.dbf
DISK INPUT 05-MAR-12 06-MAR-12 5305 3.08840752 /u1140/oradata/PRD1/arx01.dbf
SQL>The above output looks fine, but when i checked the long wait uery, it showed high percentage usage for OUTPUT streams. The below is the output:
TYPE IO_COUNT READY SHORT_WAITS LONG_WAITS LIO_PCT FILENAME
INPUT 16385 15361 0 1024 6.24961855 /u1140/oradata/PRD1/applsysd01.dbf
INPUT 16385 15361 0 1024 6.24961855 /u1140/oradata/PRD1/applsysx01.dbf
INPUT 16385 15361 0 1024 6.24961855 /u1140/oradata/PRD1/ard01.dbf
INPUT 16385 15361 0 1024 6.24961855 /u1140/oradata/PRD1/gld01.dbf
INPUT 8225 7711 0 514 6.24924012 /u1141/oradata/PRD1/rbs01.dbf
INPUT 8225 7711 0 514 6.24924012 /u1141/oradata/PRD1/rbs03.dbf
OUTPUT 10222 7662 0 2560 25.0440227 bk_87605_1_777154038
OUTPUT 10242 7678 0 2564 25.034173 bk_87606_1_777154038
AGGREGATE 40995 38433 0 2562 6.24954263
AGGREGATE 40995 38433 0 2562 6.24954263
10 rows selected.
TYPE IO_COUNT READY SHORT_WAITS LONG_WAITS LIO_PCT FILENAME
INPUT 16385 15361 0 1024 6.24961855 /u1140/oradata/PRD1/arx01.dbf
INPUT 16385 15361 0 1024 6.24961855 /u1140/oradata/PRD1/invd01.dbf
INPUT 8193 7681 0 512 6.24923715 /u1141/oradata/PRD1/rbs02.dbf
OUTPUT 10232 7669 0 2563 25.0488663 bk_87607_1_777167658
AGGREGATE 40963 38403 0 2560 6.24954227So, as per the output please suggest possible solutions to decrease the backup time, if not possible any other option.
Is there any problem with the tape device or is it an oracle issue.
Thanks
Peeks -
Need inputs on RMAN Backup n Recovery for multiple restore of clean backup
I will explain my scenario below:
We have the requirement of performance testing of our proposed production platform, for this i am asked to take RMAN backup of clean position (Say Day0 or baseline backup) and restore it multiple times for few rounds of testing. My question is after 1st restore (with set until time of day0 position) if i open the database with resetlogs, would i be able to restore same clean backup again? I am using control file in the environment instead of RMAN catalog database. What i can do to make this scenario work? I am already in preperation of setting rman catalog database.
Please suggest me the best way forward...
Many Thanks
NukeshHi Asif,
I tried to make it work but seems there are still issue with my recovery, can you have a look and point out the shortcommings. i am totally lost here
RMAN> list incarnation;
using target database control file instead of recovery catalog
List of Database Incarnations
DB Key Inc Key DB Name DB ID STATUS Reset SCN Reset Time
1 1 MYTEST 2433328848 PARENT 1 19-DEC-06
2 2 MYTEST 2433328848 PARENT 564488 04-NOV-10
3 3 MYTEST 2433328848 CURRENT 693719 06-NOV-10
RMAN> exit
TEST1DB-> rman target sys@MYTEST1
Recovery Manager: Release 10.2.0.3.0 - Production on Sun Nov 7 11:43:01 2010
Copyright (c) 1982, 2005, Oracle. All rights reserved.
target database Password:
connected to target database: MYTEST (DBID=2433328848, not open)
RMAN> reset database to incarnation 2;
using target database control file instead of recovery catalog
database reset to incarnation 2
RMAN> run {
2> restore database until scn 10000;
3> recover database until scn 10000;
4> alter database open resetlogs;
5> };
Starting restore at 07-NOV-10
allocated channel: ORA_DISK_1
channel ORA_DISK_1: sid=191 instance=MYTEST1 devtype=DISK
allocated channel: ORA_DISK_2
channel ORA_DISK_2: sid=190 instance=MYTEST1 devtype=DISK
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-03002: failure of restore command at 11/07/2010 11:45:18
RMAN-20208: UNTIL CHANGE is before RESETLOGS change
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-00558: error encountered while parsing input commands
RMAN-01009: syntax error: found ";": expecting one of: "allocate, alter, backup, beginline, blockrecover, catalog, change, c onnect, copy, convert, create, crosscheck, configure, duplicate, debug, delete, drop, exit, endinline, flashback, host, {, l ibrary, list, mount, open, print, quit, recover, register, release, replace, report, renormalize, reset, restore, resync, rm an, run, rpctest, set, setlimit, sql, switch, spool, startup, shutdown, send, show, test, transport, upgrade, unregister, va lidate"
RMAN-01007: at line 0 column 2 file: standard input
RMAN>
RMAN> show all;
RMAN configuration parameters are:
CONFIGURE RETENTION POLICY TO REDUNDANCY 1; # default
CONFIGURE BACKUP OPTIMIZATION OFF; # default
CONFIGURE DEFAULT DEVICE TYPE TO DISK;
CONFIGURE CONTROLFILE AUTOBACKUP ON;
CONFIGURE CONTROLFILE AUTOBACKUP FORMAT FOR DEVICE TYPE DISK TO '/rmanbackup/MYTEST/ora_cf%F';
CONFIGURE DEVICE TYPE DISK PARALLELISM 2 BACKUP TYPE TO BACKUPSET;
CONFIGURE DATAFILE BACKUP COPIES FOR DEVICE TYPE DISK TO 1; # default
CONFIGURE ARCHIVELOG BACKUP COPIES FOR DEVICE TYPE DISK TO 1; # default
CONFIGURE CHANNEL DEVICE TYPE DISK FORMAT '/rmanbackup/MYTEST/ora_df%t_s%s_s%p';
CONFIGURE MAXSETSIZE TO UNLIMITED; # default
CONFIGURE ENCRYPTION FOR DATABASE OFF; # default
CONFIGURE ENCRYPTION ALGORITHM 'AES128'; # default
CONFIGURE ARCHIVELOG DELETION POLICY TO NONE; # default
CONFIGURE SNAPSHOT CONTROLFILE NAME TO '/u01/app/oracle/product/10g/dbs/snapcf_MYTEST1.f'; # default
RMAN> exit
SQL> select name from v$database;
NAME
MYTEST
SQL> show parameter CONTROL_FILE_RECORD_KEEP_TIME
NAME TYPE VALUE
control_file_record_keep_time integer 7
SQL> -
Need help with Backup Recovery using Rman - Point in Time.
Hi all,
I am trying to recover my database to a certain point in time. Here are the details below.
Oracle database version is. 11G R1. on 2 Node RAC. OS is AIX.
Database name is Sales.
Due to a mistake by the Application team, the database is written over by bad data. Now I have to restore the database in to a point in time, where the database was good.
For this, i took a whole (full) RMAN backup, everything, all the archivelog files and Controlfiles as well.
After doing this, i dropped the entire database. So now everything is clean.
Now i have to restore and recover the database to this point in time.. 03/16/2011 12:45:00
Please guide.
The backups are located at.. /backup/sales/rman/
I am trying various things, but each time i get the msg..
ORA-01507: database not mounted
I understand.. the reason for this message is the controlfile does not exist.. as the database is in mount mode. But as i said.. i have dropped the database in order to proceed with entire restoration.
But i have taken a whole backup.. which also includes the controlfiles + archivefiles.
Please guide.. with proper steps and commands.Hi,
Priror to start with restore and recovery - Try to restore the control file from backups "/backup/sales/rman/"
Then further you can mount the db and further carry on with recovery (catalog the backups prior to recovery)
- Pavan Kumar N -
RMAN backup and restore for Disaster Recovery
Hi Guys,
I am very new to Oracle and have a question about RMAN backup and restore feature. I am simulating a disaster recovery scenario by having two VMs running oracle 11g, say hosta and hostb, hosta is sort of production db and the other a disaster recovery db (one that will be made primary when disaster occurs). My goal is to backup production db using RMAN and restore it on the other machine. For some reason when I restore the db on hostb, the command restores the previous backup but not the most recent one e.g. I took a backup yesterday (09/20) and applied it to the hostb that worked fine, but when I try to apply a fresh backup from today (09/21) it always picks up the old backup. Here's a dump of the screen:
Starting restore at 21-SEP-11
using channel ORA_DISK_1
channel ORA_DISK_1: starting datafile backup set restore
channel ORA_DISK_1: specifying datafile(s) to restore from backup set
channel ORA_DISK_1: restoring datafile 00001 to /oracle/app/dev/oradata/forums/s
ystem01.dbf
channel ORA_DISK_1: restoring datafile 00002 to /oracle/app/dev/oradata/forums/s
ysaux01.dbf
channel ORA_DISK_1: restoring datafile 00003 to /oracle/app/dev/oradata/forums/u
ndotbs01.dbf
channel ORA_DISK_1: restoring datafile 00004 to /oracle/app/dev/oradata/forums/u
sers01.dbf
channel ORA_DISK_1: reading from backup piece /oracle/app/dev/flash_recovery_are
a/FORUMS/backupset/o1_mf_nnnd0_TAG20110920T040950_77jx3zk7_.bkp
channel ORA_DISK_1: piece handle=/oracle/app/dev/flash_recovery_area/FORUMS/back
upset/o1_mf_nnnd0_TAG20110920T040950_77jx3zk7_.bkp tag=TAG20110920T040950
channel ORA_DISK_1: restored backup piece 1
channel ORA_DISK_1: restore complete, elapsed time: 00:01:05
Finished restore at 21-SEP-11
Notice that it reads the backup piece from this location /oracle/app/dev/flash_recovery_are
a/FORUMS/backupset/
whereas my latest backups are stored at a different location.
I am executing following steps at the RMAN prompt on hosta and hostb:
hosta (production site)
backup as compressed backupset database plus archivelog
backup current controlfile format'/oracle/oracle_bkup/ctl_%U'
backup spfile
hostb (Disaster site)
set dbid=13732232063
startup force nomount
restore spfile to '/oracle/app/dev/product/11.2.0/dbhome_1/dbs/spfileforums.ora' from '/export/home/dev/restore_db/backupset/2011_09_21/o1_mf_nnsnf_TAG20110921T114945_77ndg9ys_.bkp'
shutdown immediate
/* create a init<db_name>.ora file with SPFILE= /oracle/app/dev/product/11.2.0/dbhome_1/dbs/spfileforums.ora */
startup force pfile='/export/home/dev/restore_db/initforums.ora' nomount
restore controlfile from '/export/home/dev/restore_db/backupset/2011_09_21/ctl_1hmn3mic_1_1'
/* restart rman here */
quit
alter database mount;
catalog start with '/export/home/dev/restore_db/backupset/2011_09_21' noprompt;
/* call the next two commands on run */
restore database;
recover database;
alter database open resetlogs
quit
Any help will be greatly appreciated.
Thanks,
RajeshThanks guys, I really appreciate all your help here. I redid everything all over again to get all the information you guys wanted. Since I noticed that more eyes are looking into this I am going to reiterate my steps one more time followed by specific answers to questions. My first backup on Host B is located under ..../restore_db/backupset whereas the subsequent one is under .../restore_db/backupset/backupset2.
I take backup on Host A using:
rman target /
backup as compressed backupset database plus archivelog;
backup spfile;
quit;
I restore the backup on Host B using:
set dbid=13732232063;
startup force nomount;
restore spfile to '/oracle/app/dev/product/11.2.0/dbhome_1/dbs/spfileforums.ora' from '/export/home/dev/restore_db/backupset/o1_mf_nnsnf_TAG20110928T171830_787gbpxh_.bkp'
shutdown immediate;
startup force nomount;
restore controlfile from '/export/home/dev/restore_db/backupset/o1_mf_ncsnf_TAG20110928T171638_787gbkxn_.bkp'
quit;
/* restart rman here */
alter database mount;
catalog start with '/export/home/dev/restore_db/backupset' noprompt;
restore database;
recover database;
alter database open resetlogs;
quit;
I take another backup on Host A using (notice no spfile backup this time):
backup as compressed backupset database plus archivelog;
quit;
I restore the database on Host B using:
alter database mount;
catalog start with '/export/home/dev/restore_db/backupset/backupset2' noprompt;
recover database;
alter database open;
quit;
Output of List Backup of database (I have done this after I recovered the second time, also note that it is referring to backupset2 which is were my second backup is stored)
RMAN> list backup of database;
using target database control file instead of recovery catalog
List of Backup Sets
===================
BS Key Type LV Size Device Type Elapsed Time Completion Time
89 Full 261.87M DISK 00:01:37 28-SEP-11
BP Key: 91 Status: AVAILABLE Compressed: YES Tag: TAG20110928T171638
Piece Name: /export/home/dev/restore_db/backupset/o1_mf_nnndf_TAG2011092
8T171638_787g77rx_.bkp
List of Datafiles in backup set 89
File LV Type Ckp SCN Ckp Time Name
1 Full 1813849 28-SEP-11 /oracle/app/dev/oradata/forums/system01.dbf
2 Full 1813849 28-SEP-11 /oracle/app/dev/oradata/forums/sysaux01.dbf
3 Full 1813849 28-SEP-11 /oracle/app/dev/oradata/forums/undotbs01.dbf
4 Full 1813849 28-SEP-11 /oracle/app/dev/oradata/forums/users01.dbf
BS Key Type LV Size Device Type Elapsed Time Completion Time
97 Full 259.16M DISK 00:00:00 28-SEP-11
BP Key: 100 Status: AVAILABLE Compressed: YES Tag: TAG20110928T18352
7
Piece Name: /export/home/dev/restore_db/backupset/backupset2/o1_mf_nnndf
TAG20110928T183527787lv0nb_.bkp
List of Datafiles in backup set 97
File LV Type Ckp SCN Ckp Time Name
1 Full 1816853 28-SEP-11 /oracle/app/dev/oradata/forums/system01.dbf
2 Full 1816853 28-SEP-11 /oracle/app/dev/oradata/forums/sysaux01.dbf
3 Full 1816853 28-SEP-11 /oracle/app/dev/oradata/forums/undotbs01.dbf
4 Full 1816853 28-SEP-11 /oracle/app/dev/oradata/forums/users01.dbf
Output of list backup (done after restoring the control file)
RMAN> list backup;
BS Key Size Device Type Elapsed Time Completion Time
87 89.20M DISK 00:00:26 28-SEP-11
BP Key: 87 Status: AVAILABLE Compressed: YES Tag: TAG20110928T171526
Piece Name: /oracle/app/dev/flash_recovery_area/FORUMS/backupset/2011_09
_28/o1_mf_annnn_TAG20110928T171526_787g50bm_.bkp
List of Archived Logs in backup set 87
Thrd Seq Low SCN Low Time Next SCN Next Time
1 34 1302253 20-SEP-11 1306439 20-SEP-11
1 35 1306439 20-SEP-11 1307647 20-SEP-11
1 36 1307647 20-SEP-11 1307701 20-SEP-11
1 37 1307701 20-SEP-11 1311393 20-SEP-11
1 38 1311393 20-SEP-11 1311511 20-SEP-11
1 39 1311511 20-SEP-11 1332479 20-SEP-11
1 40 1332479 20-SEP-11 1344418 20-SEP-11
1 41 1344418 20-SEP-11 1350409 20-SEP-11
1 42 1350409 20-SEP-11 1350449 20-SEP-11
1 43 1350449 20-SEP-11 1350854 21-SEP-11
1 44 1350854 21-SEP-11 1350895 21-SEP-11
1 45 1350895 21-SEP-11 1353114 21-SEP-11
1 46 1353114 21-SEP-11 1353254 21-SEP-11
1 47 1353254 21-SEP-11 1353865 21-SEP-11
1 48 1353865 21-SEP-11 1353988 21-SEP-11
1 49 1353988 21-SEP-11 1375403 21-SEP-11
1 50 1375403 21-SEP-11 1376149 21-SEP-11
1 51 1376149 21-SEP-11 1376206 21-SEP-11
1 52 1376206 21-SEP-11 1376246 21-SEP-11
1 53 1376246 21-SEP-11 1379990 21-SEP-11
1 54 1379990 21-SEP-11 1380229 21-SEP-11
1 55 1380229 21-SEP-11 1380266 21-SEP-11
1 56 1380266 21-SEP-11 1380528 21-SEP-11
1 57 1380528 21-SEP-11 1380724 21-SEP-11
1 58 1380724 21-SEP-11 1380861 21-SEP-11
1 59 1380861 21-SEP-11 1381033 21-SEP-11
1 60 1381033 21-SEP-11 1381077 21-SEP-11
1 61 1381077 21-SEP-11 1402243 22-SEP-11
1 62 1402243 22-SEP-11 1423341 22-SEP-11
1 63 1423341 22-SEP-11 1435456 22-SEP-11
1 64 1435456 22-SEP-11 1454415 23-SEP-11
1 65 1454415 23-SEP-11 1490903 23-SEP-11
1 66 1490903 23-SEP-11 1491266 23-SEP-11
1 67 1491266 23-SEP-11 1491347 23-SEP-11
1 68 1491347 23-SEP-11 1492761 23-SEP-11
1 69 1492761 23-SEP-11 1492891 23-SEP-11
1 70 1492891 23-SEP-11 1493678 23-SEP-11
1 71 1493678 23-SEP-11 1493704 23-SEP-11
1 72 1493704 23-SEP-11 1494741 23-SEP-11
1 73 1494741 23-SEP-11 1494790 23-SEP-11
1 74 1494790 23-SEP-11 1510154 23-SEP-11
1 75 1510154 23-SEP-11 1514286 23-SEP-11
1 76 1514286 23-SEP-11 1531967 24-SEP-11
1 77 1531967 24-SEP-11 1543266 24-SEP-11
1 78 1543266 24-SEP-11 1558427 24-SEP-11
1 79 1558427 24-SEP-11 1566924 24-SEP-11
1 80 1566924 24-SEP-11 1578292 24-SEP-11
1 81 1578292 24-SEP-11 1596894 25-SEP-11
BS Key Size Device Type Elapsed Time Completion Time
88 84.03M DISK 00:00:30 28-SEP-11
BP Key: 88 Status: AVAILABLE Compressed: YES Tag: TAG20110928T171526
Piece Name: /oracle/app/dev/flash_recovery_area/FORUMS/backupset/2011_09
_28/o1_mf_annnn_TAG20110928T171526_787g63s9_.bkp
List of Archived Logs in backup set 88
Thrd Seq Low SCN Low Time Next SCN Next Time
1 82 1596894 25-SEP-11 1609028 25-SEP-11
1 83 1609028 25-SEP-11 1622303 25-SEP-11
1 84 1622303 25-SEP-11 1626430 25-SEP-11
1 85 1626430 25-SEP-11 1634486 25-SEP-11
1 86 1634486 25-SEP-11 1648398 25-SEP-11
1 87 1648398 25-SEP-11 1669259 26-SEP-11
1 88 1669259 26-SEP-11 1686820 26-SEP-11
1 89 1686820 26-SEP-11 1686959 26-SEP-11
1 90 1686959 26-SEP-11 1689168 26-SEP-11
1 91 1689168 26-SEP-11 1704759 26-SEP-11
1 92 1704759 26-SEP-11 1719597 27-SEP-11
1 93 1719597 27-SEP-11 1740407 27-SEP-11
1 94 1740407 27-SEP-11 1750125 27-SEP-11
1 95 1750125 27-SEP-11 1765592 27-SEP-11
1 96 1765592 27-SEP-11 1781498 28-SEP-11
1 97 1781498 28-SEP-11 1802311 28-SEP-11
1 98 1802311 28-SEP-11 1811009 28-SEP-11
1 99 1811009 28-SEP-11 1813811 28-SEP-11
BS Key Type LV Size Device Type Elapsed Time Completion Time
89 Full 261.87M DISK 00:01:37 28-SEP-11
BP Key: 89 Status: AVAILABLE Compressed: YES Tag: TAG20110928T171638
Piece Name: /oracle/app/dev/flash_recovery_area/FORUMS/backupset/2011_09
_28/o1_mf_nnndf_TAG20110928T171638_787g77rx_.bkp
List of Datafiles in backup set 89
File LV Type Ckp SCN Ckp Time Name
1 Full 1813849 28-SEP-11 /oracle/app/dev/oradata/forums/system01.dbf
2 Full 1813849 28-SEP-11 /oracle/app/dev/oradata/forums/sysaux01.dbf
3 Full 1813849 28-SEP-11 /oracle/app/dev/oradata/forums/undotbs01.dbf
4 Full 1813849 28-SEP-11 /oracle/app/dev/oradata/forums/users01.dbf -
BACKUP AND RECOVERY USING RMAN
제품 : ORACLE SERVER
작성날짜 : 2003-08-04
BACKUP AND RECOVERY USING RMAN
==============================
1. Backup Method
1.1 backing up in noarchivelog mode
SVRMGR> shutdown;
SVRMGR> startup mount
rman을 가동시킨 후 다음을 수행
run {
# backup the database to disk
allocate channel dev1 type disk;
backup (database format '/oracle/backups/bp_%s_%p'); }
1.2 Backing up databases and tablespaces in archivelog mode >
- database 단위 backup script
run {
allocate channel dev1 type 'sbt_tape';
backup skip offline (database format '/oracle/backups/%d_%u');
release channel dev1;
- tablespace 단위 backup script
run {
allocate channel dev1 type disk;
backup
(tablespace system,tbs_1,tbs_2,tbs_3,tbs_4,tbs_5
format '/oracle/backups/bp_%s_%p');
- datafile 단위 backup script
run {
allocate channel dev1 type disk;
backup
(datafile '?/dbs/t_dbs1.f'
format '/oracle/backups/%d_%u');
- control file backup script
control file은 system datafile의 첫번째 화일을 백업받을 때 백업되며 다음과
같은 방법으로도 백업받을 수 있다.
run {
allocate channel dev1 type 'sbt_tape';
backup
(tablespace tbs_5 include current controlfile
format '%d_%u');
- archived logs backup script
NLS_LANG=american
NLS_DATE_FORMAT='Mon DD YYYY HH24:MI:SS'
run {
allocate channel dev1 type 'sbt_tape';
backup
(archivelog from time 'Nov 13 1996 20:57:13'
until time 'Nov 13 1996 21:06:05'
all
format '%d_%u');
run {
allocate channel dev1 type 'sbt_tape';
backup
(archivelog low logseq 288 high logseq 301 thread 1
all delete input
format '%d_%u');
- copying datafiles
run {
allocate channel dev1 type disk;
copy datafile '?/dbs/tbs_01.f/dbs/tbs_01.f' to '?/copy/temp3.f';
- incremental backups
새로운 데이타화일이 추가되거나 테이블스페이스가 추가된다면 level 0의 backup을
반드시 수행한다.
run {
allocate channel dev1 type 'sbt_tape';
backup incremental level 0
(database
format '%d_%u');
run {
allocate channel dev1 type 'sbt_tape';
backup incremental level 1
(database
format '&d_%u');
2. Recovery Method
- tablespace recovery
run {
allocate channel dev1 type disk;
allocate channel dev2 type 'sbt_tape';
sql "alter tablespace tbs_1 offline immediate" ;
set newname for datafile 'disk7/oracle/tbs11.f'
to 'disk9/oracle/tbs11.f' ;
restore (tablespace tbs_1) ;
switch datafile all ;
recover tablespace tbs_1 ;
sql "alter tablespace tbs_1 online" ;
release channel dev1;
release channel dev2;
- point-in-time recovery
TBS_1 테이블스페이스는 두 개의 데이타 화일 소유. TEMP1은 백업이 없으나 user
data는 없는 temporary segemnt만 가지고 있음.
NLS_LANG=american
NLS_DATE_FORMAT='Mon DD YYYY HH24:MI:SS'
SVRMGR> shutdown abort;
SVRMGR> startup nomount;
run {
# recover database until 3pm after restoring tbs_1 to a new location
allocate channel dev1 type disk;
allocate channel dev2 type 'sbt_tape';
set until time 'Nov 15 1996 15:00:00'
set newname for datafile '/vobs/oracle/dbs/tbs_11.f'
to '?/dbs/temp1.f' ;
set newname for datafile '?/dbs/tbs_12.f'
to '?/dbs/temp2.f' ;
restore controlfile to '/vobs/oracle/dbs/cf1.f' ;
replicate controlfile from '/vobs/oracle/dbs/cf1.f';
sql "alter database mount" ;
restore database skip tablespace temp1;
switch datafile all;
recover database skip tablespace temp1;
sql "alter database open resetlogs";
sql "drop tablespace temp1";
sql "create tablespace temp1 datafile '/vobs/oracle/dbs/temp1.f' size 10M";
release channel dev1;
release channel dev2;
- 전체 데이타베이스를 복구하려 할때
current redologfile이 존재할때 다음과 같은 작업으로 전체 데이타베이스를 복구할
수 있다.
replace script restore_recover_db_cf {
execute script restore_cf;
sql 'alter database mount';
execute script restore_db;
execute script recover_db;
sql 'alter database open resetlogs';
replace script alloc_1_disk {
allocate channel d1 type disk;
setlimit channel d1 kbytes 2097150 maxopenfiles 32 readrate 200;
replace script rel_1_disk {
release channel d1;
replace script restore_cf {
execute script alloc_1_disk;
restore
controlfile to '/private/db_files/twih/dbf/ctrl_twih_1.ctl';
replicate
controlfile from '/private/db_files/twih/dbf/ctrl_twih_1.ctl';
execute script rel_1_disk;
replace script restore_db {
execute script alloc_1_disk;
restore
(database);
execute script rel_1_disk;
replace script recover_db {
execute script alloc_1_disk;
recover
database;
execute script rel_1_disk;
Reference Documents
<Note:104796.1>Hi Alok,
why there is a need for standby here? you mean to say standby db?
what i exactly need is , whether i will be making use of archivelogs which are all archived during that backup period (in mount stage) after some days, if i face any problem with the PRODUCTION Database.. -
What are the steps in RMAN Backup & Recovery 10 G
Hi All
Please list out What are the steps in RMAN Backup & Recovery 10 G
RegardsTake a look at the user guide or at the getting started section, dba core tasks - backup and recovery.
C. -
How to estimate recovery time from rman backup
how to estimate recovery time from rman backup?
The database is of size 800GB.
the approximate time to backup is 8- 10 hours."Recovery time" is : RESTORE DATABASE time + RECOVER DATABASE time.
RESTORE can be done
a. in parallel with multiple channels if your backup was created as multiple backupsets
b. using incremental backups
c. whether the backups are on
i. disk
ii. tape
Alternatively, if you have a Recovery Area with an Incrementally Update backup of the database, you can SWITCH DATABASE almost immediately
RECOVER depends on
a. how many archivelogs (in terms of size, not necessarily number) need to be applied
-- this depends on
i. transaction volume
ii. the "freshness" of the backups (how recent the backups are)
b. whether required archivelogs are available on disk (which can be the case if the database backup itself is a very recent backup and archivelogs since then haven't been deleted {whether they have been backed up or not}) OR whether they need to be restored from
i. disk backup
ii. tape backup -
SQL to find rman backup files required for recovery?
Hi folks,
I'm looking for a sql statement that I can use to query the database and return a list of the RMAN backup files needed for a successful recovery based on a SCN or a point in time. I want to do it through SQL (not though rman) because I want to script an automated refresh process. My databases are backed up online, so I would need any archive log backups done during the database backup as well. So for example...
I do a database backup
Then do an archive log backup.
Pass a SQL script the lowest and highest scn numbers during the backup.
It returns a list of the rman backup pieces needed to do a successful recovery.
I've got a script, below, which shows the lowest and highest scn during the last full backup (all datafiles which exist have been backed up), if someone understands what I'm looking for and has done something similar, I would really appreciate any insight you may have. What I'd want to get back from the SQL is the names of the rman backup files (handles) which would be needed to get the database recovered to the highest scn during the backup.
SELECT MIN(VBD.CHECKPOINT_CHANGE#), MAX(VBD.CHECKPOINT_CHANGE#)
FROM V$BACKUP_DATAFILE VBD, V$DATAFILE VDF,
(SELECT
CREATION_CHANGE#,
MAX(COMPLETION_TIME) COMPLETION_TIME
FROM V$BACKUP_DATAFILE
WHERE CREATION_CHANGE# IN (
SELECT CREATION_CHANGE# FROM V$DATAFILE)
GROUP BY CREATION_CHANGE#
) QUERY1
WHERE VBD.CREATION_CHANGE# = VDF.CREATION_CHANGE#
AND VBD.CREATION_CHANGE# = QUERY1.CREATION_CHANGE#
AND VBD.COMPLETION_TIME = QUERY1.COMPLETION_TIME;
Thanks very much for your time,
-Adam vonNiedaIn case anyone's wondering, here's the answer. The following query returns the min and max SCN numbers contained in the last online backup.
SELECT MIN(VBD.CHECKPOINT_CHANGE#), MAX(VBD.CHECKPOINT_CHANGE#)
FROM V$BACKUP_DATAFILE VBD, V$DATAFILE VDF,
(SELECT
CREATION_CHANGE#,
MAX(COMPLETION_TIME) COMPLETION_TIME
FROM V$BACKUP_DATAFILE
WHERE CREATION_CHANGE# IN (
SELECT CREATION_CHANGE# FROM V$DATAFILE)
GROUP BY CREATION_CHANGE#
) QUERY1
WHERE VBD.CREATION_CHANGE# = VDF.CREATION_CHANGE#
AND VBD.CREATION_CHANGE# = QUERY1.CREATION_CHANGE#
AND VBD.COMPLETION_TIME = QUERY1.COMPLETION_TIME;
This next one gets the session key for that backup.
Select distinct session_key from v$backup_datafile_details
where checkpoint_change# >= <min scn# from above> and checkpoint_change# <= <max scn# from above>;
And finally, this one returns all files written out during that RMAN session, which includes control files, data files, and archived redo logs
Select handle from v$backup_piece_details
where session_key = <session key from above> order by completion_time;
Maybe you are looking for
-
i purchased a song on my iphone and it is not showing up in my library on my computer. How do I get this on here so that i can put it on my daughters ipod.?
-
Please add a no-sign-in feature for Muse in Enterprise
We install Muse in a multimedia lab for our university, in order to provide a wonderfully easy web design tool. Our students and faculty are put off and confused by the login screen and the inconvenience of needing to create an adobe ID before access
-
Watch keep failing heart rate detection
Does anyone has problem with the heart rate monitor? I have normal skin color (light, no tattoos) and the heart rate monitor can't find my pulse. The last correct one it detected was 24 hours ago. It's been failing for more than 10 times in a row and
-
How to save a user version of interactive reports in 3.2.
My users ask me to save a "modified version" of a defined interactive report (eg. the want to change the selected columns and the remember them). When I connect to a report as developer I see a "Save Report *" option which asks me if I want to apply
-
When I try to see YouTube, I am told that the server is not correctly configured. A pc on the same wifi network has no problems. What am I doing wrong?