Category: RMAN Restore and Recovery

Create a 19c RAC Standby Database From Overnight RMAN Tape Backups

Here is an example of building a RAC standby database by using overnight  NetBackup RMAN backups.

Primary RAC servers: racnode1,racnode2
Standby RAC servers: stdbynode1,stdbynode2
Primary Database Name:  RACTEST
Primary Instance Names: RACTEST1/2
Standby Instance Names: STBTEST1/2

RAC Database version: 19.x.x.x.x

1)On primary database, check most recent RMAN backups are available.

$ rman target /  catalog rman/password@rman
RMAN> list backup of database;

STEP 3 and 4 are not required, if using DUPLICATE.

2) On primary, create a standby controlfile.

SQL> alter database create standby controlfile as '/tmp/RACTEST_standby.ctl';
Database altered.

or

RMAN>  backup device type disk format '/tmp/RACTEST_standby.ctl' current controlfile for standby;

3)Copy standby controlfile onto standby server stdbynode1:/tmp.

[RACTEST1] oracle@racnode1:/tmp$ scp RACTEST_standby.ctl stdbynode1:/tmp/

-4) Copy primary database password file onto standby database server as per “How to copy password file from primary to standby database in 12c onwards“

5) On primary database, make change for FORCE LOGGING, only if database is not in “force logging” yet.

 SQL> ALTER DATABASE FORCE LOGGING;

6)  On primary database,  change or add below parameters .

DB_UNIQUE_NAME=RACTEST
LOG_ARCHIVE_CONFIG='DG_CONFIG=(STBTEST,RACTEST)'
DB_FILE_NAME_CONVERT='STBTEST','RACTEST'
LOG_FILE_NAME_CONVERT='STBTEST','RACTEST'
STANDBY_FILE_MANAGEMENT=AUTO
FAL_SERVER=STBTEST
FAL_CLIENT=RACTEST
LOG_ARCHIVE_DEST_STATE_1=ENABLE
LOG_ARCHIVE_DEST_STATE_2=ENABLE
log_archive_dest_1='LOCATION=use_db_recovery_file_dest VALID_FOR=( ALL_LOGFILES, ALL_ROLES) DB_UNIQUE_NAME=RACTEST'
LOG_ARCHIVE_DEST_2='SERVICE=STBTEST VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=STBTEST'

7) On standby database, change or add below parameters.

DB_NAME=RACTEST
DB_UNIQUE_NAME=STBTEST
LOG_ARCHIVE_CONFIG='DG_CONFIG=(RACTEST,STBTEST)'
DB_FILE_NAME_CONVERT='RACTEST','STBTEST'
LOG_FILE_NAME_CONVERT='RACTEST','STBTEST'
STANDBY_FILE_MANAGEMENT=AUTO
FAL_SERVER=RACTEST
FAL_CLIENT=STBTEST
*.log_archive_dest_1='LOCATION=use_db_recovery_file_dest VALID_FOR=( ALL_LOGFILES, ALL_ROLES) DB_UNIQUE_NAME=STBTEST'
*.LOG_ARCHIVE_DEST_2='SERVICE=RACTEST LGWR SYNC VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=RACTEST'

8) On standby database, set standby *.control_files=(‘+DATA1′,’+FRA’), and startup database in nomount mode.

SQL> startup nomount;
ORACLE instance started.

Total System Global Area 1.5366E+10 bytes
Fixed Size 2165488 bytes
Variable Size 7650411792 bytes
Database Buffers 7650410496 bytes
Redo Buffers 63135744 bytes

9) On standby database, restore standby control files. Update parameter “control_files” accordingly.

No need this step if using DUPLICATE.

[STBTEST1] oracle@stdbynode1:/tmp$ rman target /
RMAN> restore controlfile to '+FRA/STBTEST/CONTROLFILE/controlfile2.f' from '/tmp/RACTEST2_standby.ctl';
RMAN> restore controlfile to '+DATA1/STBTEST/CONTROLFILE/controlfile1.f'  from '/tmp/RACTEST2_standby.ctl';
control_files = '+DATA1/STBTEST/CONTROLFILE/controlfile1.f','+fra/STBTEST/CONTROLFILE/controlfile2.f'

10) On standby database, shutdown and mount standby database.

SQL>  create spfile='+DATA1/STBTEST/PARAMETERFILE/spfileSTBTEST.ora' from pfile='/tmp/stbtest.ora';

File created.

$  view $ORACLE_HOME/dbs/initSTBTEST1.ora
$ cat $ORACLE_HOME/dbs/initSTBTEST1.ora
spfile='+DATA1/STBTEST/PARAMETERFILE/spfileSTBTEST.ora' 


SQL> shutdown immediate;
ORA-01507: database not mounted


ORACLE instance shut down.

SQL> startup nomount;
ORACLE instance started.
Total System Global Area 1.0262E+11 bytes
Fixed Size 2265224 bytes
Variable Size 1.2885E+10 bytes
Database Buffers 8.9657E+10 bytes
Redo Buffers 74416128 bytes

-- no need the following steps if using DUPLICATE
--

SQL> alter database mount standby database;
Database altered.

11) On standby database, restore datafiles from overnight RMAN backups.

[STBTEST1] oracle@stdbynode1:/u01/app/oracle/admin/STBTEST/scripts$ cat restore_STBTEST.rman

run {
allocate channel c1 device type sbt_tape parms='ENV=(NB_ORA_CLIENT=racnode1)';
allocate channel c2 device type sbt_tape parms='ENV=(NB_ORA_CLIENT=racnode1)';
allocate channel c3 device type sbt_tape parms='ENV=(NB_ORA_CLIENT=racnode1)';
allocate channel c4 device type sbt_tape parms='ENV=(NB_ORA_CLIENT=racnode1)';
SET NEWNAME FOR DATABASE TO '+DATA1';
restore database;
switch datafile all;
switch tempfile all;
release channel c1;
release channel c2;
release channel c3;
release channel c4;
}

[STBTEST1] oracle@stdbynode1:/u01/app/oracle/admin/STBTEST/scripts$ nohup rman target / catalog rman/password@rman cmdfile=/u01/app/oracle/admin/STBTEST/scripts/restore_STBTEST.rman log=/u01/app/oracle/admin/STBTEST/scripts/restore_STBTEST.log &
[1] 19448

By using “restore database”, you have to add “switch datafile all” , because the ASM disk path for datafiles and redolog files have changed and they are actually from ones in standby control file, and otherwise the datafile name will be like:

*****************************************
WARNING: The converted filename '+REDO/STBTEST/onlinelog/group_2.280.1163406521'
         is an ASM fully qualified filename.
         Changing the filename to '+REDO/MUST_RENAME_THIS_LOGFILE_2.4294967295.4294967295'.
         Please rename it accordingly.
*****************************************
*****************************************
WARNING: The converted filename '+DATA1/STBTEST/datafile/users.854.1163407487'
        is an ASM fully qualified filename.
         Changing the filename to '+DATA1/MUST_RENAME_THIS_DATAFILE_7.4294967295.4294967295'.
         Please rename it accordingly.
*****************************************

For “MUST_RENAME_THIS_TEMPFILE“, it can be ignored, cos when we “alter database open read only” and “alter pluggable database pdb1 open read only”, those TEMP files are created automatically.

PDB1(3):*****************************************
PDB1(3):WARNING: The converted filename '+DATA1/xxx/34cbb1a7a1fcd520e0630c2101erwdd/tempfile/temp.2591.1175684807'
PDB1(3):         is an ASM fully qualified filename.
PDB1(3):         Changing the filename to '+DATA1/MUST_RENAME_THIS_TEMPFILE_21.4294967295.4294967295'.
PDB1(3):         Please rename it accordingly.

After database and PDB opened read only, DBA can drop and add temp files like :

SQL>  alter tablespace temp drop tempfile 14;

Tablespace altered.


SQL> alter tablespace temp add tempfile '+data1' size 100m autoextend on maxsize 32767m;

Tablespace altered.

OR we can

1)Change Standby_File_Management parameter to MANUAL
2) Cleaning and creating Standby Logfiles.
SQL>ALTER DATABASE DROP LOGFILE GROUP 11
…
..

SQL>ALTER DATABASE ADD STANDBY LOGFILE THREAD 1 group 11 size 1024m;
3)Cleaning and creating Online Logfiles.
SQL>ALTER DATABASE DROP LOGFILE GROUP 1;
SQL>ALTER DATABASE ADD LOGFILE THREAD 1 group 1 size 1024M

4) Cleaning and creating Temp Datafiles.
SQL> ALTER DATABASE RENAME FILE ‘+DATA/MUST_RENAME_THIS_TEMPFILE_1.4294967295.4294967295’ to ‘+DATA/STEST/TEMPFILE/tmpfile_01.dat’;

5) Change Standby_File_Management parameter to AUTO
SQL> alter system set standby_file_management=’AUTO’;

we prefer the following way to duplicate a standby database without much manual works.

[STBTEST1] oracle@stdbynode1:/u01/app/oracle/admin/STBTEST/scripts$ cat duplicate_STBTEST.rman

run {
allocate auxiliary channel c1 device type sbt_tape parms='ENV=(NB_ORA_CLIENT=racnode1)';
allocate auxiliary channel c2 device type sbt_tape parms='ENV=(NB_ORA_CLIENT=racnode1)';
allocate auxiliary channel c3 device type sbt_tape parms='ENV=(NB_ORA_CLIENT=racnode1)';
allocate auxiliary channel c4 device type sbt_tape parms='ENV=(NB_ORA_CLIENT=racnode1)';
SET NEWNAME FOR DATABASE TO '+DATA1';
DUPLICATE DATABASE RACTEST FOR STANDBY UNTIL TIME "TO_DATE('19-JAN-2025 14:32:00','DD-MON-YYYYHH24:MI:SS')";
}
release channel c1;
release channel c2;
release channel c3;
release channel c4;
}

[STBTEST1] oracle@stdbynode1:/u01/app/oracle/admin/STBTEST/scripts$ nohup rman target / catalog rman/password@rman cmdfile=/u01/app/oracle/admin/STBTEST/scripts/duplicate_STBTEST.rman log=/u01/app/oracle/admin/STBTEST/scripts/duplicate_STBTEST.log &
[1] 19449

12) On standby database, monitor the progressing by querying gv$session_longops.

SQL> select INST_ID,OPNAME,SOFAR*8/1024/ELAPSED_SECONDS,ELAPSED_SECONDS/60,TIME_REMAINING/60 from gv$session_longops where OPNAME like 'RMAN%' and TIME_REMAINING>6 order by 2;
 INST_ID OPNAME SOFAR*8/1024/ELAPSED_SECONDS ELAPSED_SECONDS/60 TIME_REMAINING/60
---------- ---------------------------------------------------------------- ---------------------------- ------------------ -----------------
 1 RMAN: aggregate input 425.803241 20.1666667 102.85
 1 RMAN: full datafile restore 105.103955 21.55 103.016667
 1 RMAN: full datafile restore 112.092109 21.9 94.7833333
 1 RMAN: full datafile restore 113.487487 21.8833333 93.4333333
 1 RMAN: full datafile restore 108.213722 21.8833333 99.35

13) Restore gap archivelogs.

RMAN>restore archivelog from logseq=100 until logseq=200 thread=1;

RMAN>restore archivelog from logseq=80 until logseq=180 thread=2;

14)  Add standby online logfiles ( n + 1 ) into both primary and standby database.

SQL> alter database add standby logfile thread 1 group 11 size 1024m;

Database altered.
....
...
..
.

15) On standby database,  create spfile if not yet.

SQL> create spfile='+DATA1' from pfile='/u01/app/oracle/product/12.1.0/dbhome_1/dbs/initSTBTEST1.ora';

File created.

$ echo spfile='+DATA1/STBTEST/PARAMETERFILE/spfile.1323.925567677' > initSTBTEST1.ora

$ scp initSTBTEST1.ora stdbynode2:/u01/app/oracle/product/12.1.0/dbhome_1/dbs/initSTBTEST2.ora

16) On standby database, put standby database under recovery.

SQL> alter database recover managed standby database cancel;

Database altered.

SQL> shutdown immediate;
ORA-01109: database not open
Database dismounted.
ORACLE instance shut down.

SQL> startup nomount;


ORACLE instance started.

Total System Global Area 1.0489E+10 bytes
Fixed Size 2162560 bytes
Variable Size 5704253568 bytes
Database Buffers 4764729344 bytes
Redo Buffers 17899520 bytes

SQL> alter database mount standby database;

Database altered.

SQL> alter database recover managed standby database disconnect from session;

Database altered.

SQL> alter database recover managed standby database disconnect nodelay;

17) Add standby database into cluster.

[STBTEST1] oracle@stdbynode1:/u01/app/oracle/product/12.1.0/dbhome_1/dbs$ srvctl add database -n STBTEST -o /u01/app/oracle/product/12.1.0/dbhome_1 -d STBTEST -r physical_standby
[STBTEST1] oracle@stdbynode1:/u01/app/oracle/product/12.1.0/dbhome_1/dbs$ srvctl modify database -d STBTEST -s mount
[STBTEST1] oracle@stdbynode1:/u01/app/oracle/product/12.1.0/dbhome_1/dbs$ srvctl add instance -d STBTEST -i STBTEST1 -n stdbynode1
[STBTEST1] oracle@stdbynode1:/u01/app/oracle/product/12.1.0/dbhome_1/dbs$ srvctl add instance -d STBTEST -i STBTEST2 -n stdbynode2

18) Configure archivelog deletion policy for both primary and standby databases.

a) On primary, log on to RMAN and change the archive log deletion policy by running:

RMAN>CONFIGURE ARCHIVELOG DELETION POLICY TO NONE;

b) On standby, log on to RMAN and change the archive log deletion policy by running:

RMAN>CONFIGURE ARCHIVELOG DELETION POLICY TO APPLIED ON STANDBY;

19) For 12c onwards, no need to add a static registration into LISTENER.ora on all nodes of both primary and standby database. Below is only for under 12c like 11.2.0.4, etc.

#
# DGMGRL>switchover to <standby> Fails with ORA-12514 (Doc ID 1582927.1)
# Each Standby needs a static registration for the switchover to work
#
-- for node racnode1

SID_LIST_LISTENER =
 (SID_LIST =
 (SID_DESC =
 (GLOBAL_DBNAME = STBTEST_DGMGRL)
 (ORACLE_HOME = /u01/app/oracle/product/11.2.0/dbhome_1)
 (SID_NAME = RACTEST1)
 )
 )

-- for node racnode2

SID_LIST_LISTENER =
 (SID_LIST =
 (SID_DESC =
 (GLOBAL_DBNAME = STBTEST_DGMGRL)
 (ORACLE_HOME = /u01/app/oracle/product/11.2.0/dbhome_1)
 (SID_NAME = RACTEST2)
 )
 )

-- for node stdbynode1

SID_LIST_LISTENER =
 (SID_LIST =
 (SID_DESC =
 (GLOBAL_DBNAME = STBTEST_DGMGRL)
 (ORACLE_HOME = /u01/app/oracle/product/11.2.0/dbhome_1)
 (SID_NAME = STBTEST1)
 )
 )

-- for node stdbynode2

SID_LIST_LISTENER =
 (SID_LIST =
 (SID_DESC =
 (GLOBAL_DBNAME = STBTEST_DGMGRL)
 (ORACLE_HOME = /u01/app/oracle/product/11.2.0/dbhome_1)
 (SID_NAME = STBTEST2)
 )
 )


DGMGRL> show instance 'STBTEST1' StaticConnectIdentifier
 StaticConnectIdentifier = '(DESCRIPTION=(ADDRESS=(PROTOCOL=TCP)(HOST=10.10.10.10)(PORT=1521))(CONNECT_DATA=(SERVICE_NAME=STBTEST_DGMGRL)(INSTANCE_NAME=STBTEST1)(SERVER=DEDICATED)))'
DGMGRL> show instance 'STBTEST2' StaticConnectIdentifier
 StaticConnectIdentifier = '(DESCRIPTION=(ADDRESS=(PROTOCOL=TCP)(HOST=10.10.10.12)(PORT=1521))(CONNECT_DATA=(SERVICE_NAME=STBTEST_DGMGRL)(INSTANCE_NAME=STBTEST2)(SERVER=DEDICATED)))'
DGMGRL> show instance 'RACTEST1' StaticConnectIdentifier
 StaticConnectIdentifier = '(DESCRIPTION=(ADDRESS=(PROTOCOL=TCP)(HOST=10.10.10.20)(PORT=1521))(CONNECT_DATA=(SERVICE_NAME=RACTEST_DGMGRL)(INSTANCE_NAME=RACTEST1)(SERVER=DEDICATED)))'
DGMGRL> show instance 'RACTEST2' StaticConnectIdentifier
 StaticConnectIdentifier = '(DESCRIPTION=(ADDRESS=(PROTOCOL=TCP)(HOST=10.10.10.22)(PORT=1521))(CONNECT_DATA=(SERVICE_NAME=RACTEST_DGMGRL)(INSTANCE_NAME=RACTEST2)(SERVER=DEDICATED)))'

Create and enable DG broker

set db_broker_config_file1 and 2 on both primary and standby. remove ‘LOG_ARCHIVE_DEST_2’ on standby 

-- on standby only
SQL> alter system set LOG_ARCHIVE_DEST_2='';
SQL> alter system set dg_broker_config_file1='+DATA2/RACTEST/DATAGUARDCONFIG/dr1RACTEST.dat';

System altered.

SQL> alter system set dg_broker_config_file2='+DATA2/RACTEST/DATAGUARDCONFIG/dr2RACTEST.dat';

System altered.

SQL> alter system set dg_broker_start=true;
DGMGRL> connect sys
Password:
Connected to "RACTEST"
Connected as SYSDBA.
DGMGRL> CREATE CONFIGURATION DG_RACTEST AS PRIMARY DATABASE IS RACTEST CONNECT IDENTIFIER IS RACTEST;
Configuration "dg_ractest" created with primary database "ractest"

DGMGRL> ADD DATABASE STBTEST AS CONNECT IDENTIFIER is STBTEST MAINTAINED AS PHYSICAL;
Database "stbtest" added

System altered.


DGMGRL> enable configuration;

DGMGRL> show configuration;

RMAN-06023 Although Backups Are Available 

‘RMAN  RESTORE DATABASE’ command fails with the following errors by using a backup control file:

RMAN-00571: ====================================================       
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ========       
RMAN-00571: ====================================================       
RMAN-03002: failure of restore command at 21/11/2024 11:09:9           
RMAN-06026: some targets not found - aborting restore                  
....
RMAN-06023: no backup or copy of datafile 3 found to restore           
RMAN-06023: no backup or copy of datafile 2 found to restore           
RMAN-06023: no backup or copy of datafile 1 found to restore

If you run ‘list backup of database’, we can see the database backups available.

REASON

The problem here is that there are some files in the Flash Recovery Area that belong to different incarnation than the available backups CURRENT incarnation.

If we start a RESTORE database with a BACKUP controlfile and Flash Recovery Area is defined, RMAN execute and implicit crosscheck and catalog of all the objects in the Flash Recovery Area.

SOLOTION

If the offending cataloged files are small, we can move those files to a directory outside the Flash Recovery Area. Then we need to restart the whole restore procedure. It’s necessary to restore controlfile again

OR

Another solution is to temporarily disable use of the flash recovery area for the duration of the restore and recovery commands.

    To disable Flash Recovery Area you need to undefine db_recovery_file_dest:

    #*.db_recovery_file_dest='<directory>'
    #*.db_recovery_file_dest_size=<size>

    Bounce database

    Restore controlfile and then restore/recover database commands.

    If there are some backuppieces or archivelogs in the Flash Recovery Area that need to be cataloged, then it will be necessary to catalog them manually with:  CATALOG BACKUPPIECE  or CATALOG ARCHIVELOG commands.

    Create CDB Error: Signature of package “SYS.DBMS_BACKUP_RESTORE” has been changed

    The following errors occur when creating a CDB database with patch level 19.23.

    ERROR=RMAN-00571: =======================================================
    ERROR=RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS 
    ERROR=RMAN-00571: ============================================
    ERROR=ORA-04062: signature of package "SYS.DBMS_BACKUP_RESTORE" has been changed
    SEVERE: [FATAL] Error while restoring PDB backup piece
    oracle.sysman.assistants.util.step.StepExecutionException: Error while restoring PDB backup piece

    WORKAROUND

    As a workaround, by selecting the “Custom database” option instead of “General Purpose Database”.

    PL/SQL package SYS.DBMS_BACKUP_RESTORE version PL/SQL package SYS.DBMS_RCVMAN version in TARGET database is not current

    Try to connect to RMAN catalog with the following errors:

    PL/SQL package SYS.DBMS_BACKUP_RESTORE version 19.20.00.00 in TARGET database is not current
    PL/SQL package SYS.DBMS_RCVMAN version 19.20.00.00 in TARGET database is not current
    connected to target database: TESTDB (DBID=1231118758)
    connected to recovery catalog database

    CAUSE

    Once the DBRU/DBRUR is applied  datapatch was not run. While DBRU/DBRUR will update the binary datpatch would ensure the dictionary are also updated

    SOLUTION

    Crosscheck if datapatch was run.

    SET PAGESIZE 100
    SET LINESIZE 300
    COLUMN status FORMAT A10
    COLUMN description FORMAT A40
    COLUMN source_version FORMAT A10
    COLUMN target_version FORMAT A10

    select CON_ID,
    TO_CHAR(action_time, 'DD-MON-YYYY HH24:MI:SS') AS action_time,
    PATCH_ID,
    PATCH_TYPE,
    ACTION,
    DESCRIPTION,
    SOURCE_VERSION,
    TARGET_VERSION
    from CDB_REGISTRY_SQLPATCH
    order by CON_ID, action_time, patch_id;

    Run datapatch if it was not run:

    $ORACLE_HOME/OPatch/datapatch -verbose

    select * from pdb_plug_in_violations where status <> 'RESOLVED';

    $ORACLE_HOME/perl/bin/perl $ORACLE_HOME/rdbms/admin/catcon.pl -n 1 -e -b utlrp -d $ORACLE_HOME/rdbms/admin utlrp.sq

    RMAN-06136: ORACLE error from auxiliary database: ORA-00201: control file version 11.2.0.4.0 incompatible with ORACLE version 11.2.0.0.0

    1. Check parameter ‘compatible’ on both source and target databases.
    2. Set compatible value the same on target as on the source.
    3. Rerun RMAN duplicate database again.