In Oracle 11gr2, RMAN can create a duplicate database using two methods

1. Directly from the source database – If the nodes are across the network, then there will be network latency. Also, this can cause I/O issues on the source due to the read work for the RMAN copy.

2. Another method is to create duplicate database from the RMAN backup of the source. This way the clone process does not have to make any connection to the source database at all.
This is useful for big databases where the cloned environment is not in the same cluster and the source database is huge in size. The RMAN backup of the source database can be taken on the local media server using NFS mount, copied over across the network. This can be used to create the cloned environment.

In this example, I have created a clone of a two nodes database, RAC. Two instances are RAC1 and RAC2 residing on nodes rac1 and rac2 respectively. This has been cloned as DUP with instances as DUP1 and DUP2 on two nodes respectively.

Step 1: Create a password file

[oracle@rac1 rac]$ orapwd file=/u01/app/oracle/product/11.2.0/db_1/dbs/orapwDUP entries=25

Enter password for SYS:

Step 2 Create pfile for the cloned environment
SQL> create pfile=’/u01/app/oracle/product/11.2.0/db_1/dbs/initDUP.ora’ from spfile ;

File created.

Step 3: Modify the init file to comment out all cluster related parameters viz. cluster_database, instance_number, instance_name etc.

Step 4: Create folders for the database DUP under ORACLE_BASE as neededed as per the parameter file.

Step 5:Start this as auxiliary instance in nomounT
RMAN> connect auxiliary / ;

connected to auxiliary database: DUP (not mounted)

Step 6: Run the duplicate database command from RMAN

RMAN>
DUPLICATE DATABASE TO DUP
UNTIL TIME “TO_DATE(’06/06/2010 11:51:00′, ‘MM/DD/YYYY HH24:MI:SS’)”
SPFILE
BACKUP LOCATION ‘/u01/rman/rac’
NOFILENAMECHECK;

RMAN> connect auxiliary /

connected to auxiliary database: DUP (not mounted)

RMAN> DUPLICATE DATABASE TO DUP
2> UNTIL TIME “TO_DATE(’06/06/2010 11:51:00′, ‘MM/DD/YYYY HH24:MI:SS’)”
3> BACKUP LOCATION ‘/u01/rman/rac’
4> NOFILENAMECHECK;

Starting Duplicate Db at 06-JUN-10

contents of Memory Script:
{
sql clone “create spfile from memory”;
}
executing Memory Script

sql statement: create spfile from memory

contents of Memory Script:
{
shutdown clone immediate;
startup clone nomount;
}
executing Memory Script

Oracle instance shut down

connected to auxiliary database (not started)
connected to auxiliary database (not started)
Oracle instance started

Total System Global Area 849530880 bytes

Fixed Size 1339824 bytes
Variable Size 503320144 bytes
Database Buffers 339738624 bytes
Redo Buffers 5132288 bytes

contents of Memory Script:
{
sql clone “alter system set control_files =
”+DATA/dup/controlfile/current.314.720985707” comment=
”Set by RMAN” scope=spfile”;
sql clone “alter system set db_name =
”RAC” comment=
”Modified by RMAN duplicate” scope=spfile”;
sql clone “alter system set db_unique_name =
”DUP” comment=
”Modified by RMAN duplicate” scope=spfile”;
shutdown clone immediate;
startup clone force nomount
restore clone primary controlfile from ‘/u01/rman/rac/RAC_DB_0klfi2kv_20_1′;
alter clone database mount;
}
executing Memory Script

sql statement: alter system set control_files = ”+DATA/dup/controlfile/current.314.720985707” comment= ”Set by RMAN” scope=spfile

sql statement: alter system set db_name = ”RAC” comment= ”Modified by RMAN duplicate” scope=spfile

sql statement: alter system set db_unique_name = ”DUP” comment= ”Modified by RMAN duplicate” scope=spfile

Oracle instance started

Total System Global Area 849530880 bytes

Fixed Size 1339824 bytes
Variable Size 503320144 bytes
Database Buffers 339738624 bytes
Redo Buffers 5132288 bytes

Starting restore at 06-JUN-10
allocated channel: ORA_AUX_DISK_1
channel ORA_AUX_DISK_1: SID=28 device type=DISK

channel ORA_AUX_DISK_1: restoring control file
channel ORA_AUX_DISK_1: restore complete, elapsed time: 00:00:07
output file name=+DATA/dup/controlfile/current.314.720985707
Finished restore at 06-JUN-10

database mounted
released channel: ORA_AUX_DISK_1
allocated channel: ORA_AUX_DISK_1
channel ORA_AUX_DISK_1: SID=28 device type=DISK

contents of Memory Script:
{
set until scn 5582084;
set newname for clone datafile 1 to new;
set newname for clone datafile 2 to new;
set newname for clone datafile 3 to new;
set newname for clone datafile 4 to new;
set newname for clone datafile 5 to new;
set newname for clone datafile 6 to new;
restore
clone database
;
}
executing Memory Script

executing command: SET until clause

executing command: SET NEWNAME

executing command: SET NEWNAME

executing command: SET NEWNAME

executing command: SET NEWNAME

executing command: SET NEWNAME

executing command: SET NEWNAME

Starting restore at 06-JUN-10
using channel ORA_AUX_DISK_1

channel ORA_AUX_DISK_1: starting datafile backup set restore
channel ORA_AUX_DISK_1: specifying datafile(s) to restore from backup set
channel ORA_AUX_DISK_1: restoring datafile 00001 to +DATA
channel ORA_AUX_DISK_1: restoring datafile 00002 to +DATA
channel ORA_AUX_DISK_1: restoring datafile 00003 to +DATA
channel ORA_AUX_DISK_1: restoring datafile 00004 to +DATA
channel ORA_AUX_DISK_1: restoring datafile 00005 to +DATA
channel ORA_AUX_DISK_1: restoring datafile 00006 to +DATA
channel ORA_AUX_DISK_1: reading from backup piece /u01/rman/rac/RAC_DB_0jlfi2c1_19_1
channel ORA_AUX_DISK_1: piece handle=/u01/rman/rac/RAC_DB_0jlfi2c1_19_1 tag=TAG20100606T114609
channel ORA_AUX_DISK_1: restored backup piece 1
channel ORA_AUX_DISK_1: restore complete, elapsed time: 00:06:33
Finished restore at 06-JUN-10

contents of Memory Script:
{
switch clone datafile all;
}
executing Memory Script

datafile 1 switched to datafile copy
input datafile copy RECID=7 STAMP=720986311 file name=+DATA/dup/datafile/system.323.720985927
datafile 2 switched to datafile copy
input datafile copy RECID=8 STAMP=720986311 file name=+DATA/dup/datafile/sysaux.321.720985917
datafile 3 switched to datafile copy
input datafile copy RECID=9 STAMP=720986311 file name=+DATA/dup/datafile/undotbs1.319.720985933
datafile 4 switched to datafile copy
input datafile copy RECID=10 STAMP=720986312 file name=+DATA/dup/datafile/users.324.720985937
datafile 5 switched to datafile copy
input datafile copy RECID=11 STAMP=720986312 file name=+DATA/dup/datafile/example.320.720985935
datafile 6 switched to datafile copy
input datafile copy RECID=12 STAMP=720986312 file name=+DATA/dup/datafile/undotbs2.322.720985935

contents of Memory Script:
{
set until time “TO_DATE(’06/06/2010 11:51:00′, ‘MM/DD/YYYY HH24:MI:SS’)”;
recover
clone database
delete archivelog
;
}
executing Memory Script

executing command: SET until clause

Starting recover at 06-JUN-10
using channel ORA_AUX_DISK_1

starting media recovery

channel ORA_AUX_DISK_1: starting archived log restore to default destination
channel ORA_AUX_DISK_1: restoring archived log
archived log thread=1 sequence=114
channel ORA_AUX_DISK_1: restoring archived log
archived log thread=2 sequence=83
channel ORA_AUX_DISK_1: reading from backup piece /u01/rman/rac/RAC_DB_0llfi2ld_21_1
channel ORA_AUX_DISK_1: piece handle=/u01/rman/rac/RAC_DB_0llfi2ld_21_1 tag=TAG20100606T115108
channel ORA_AUX_DISK_1: restored backup piece 1
channel ORA_AUX_DISK_1: restore complete, elapsed time: 00:00:02
archived log file name=+DATA/dup/archivelog/2010_06_06/thread_1_seq_114.316.720986321 thread=1 sequence=114
archived log file name=+DATA/dup/archivelog/2010_06_06/thread_2_seq_83.271.720986321 thread=2 sequence=83
channel clone_default: deleting archived log(s)
archived log file name=+DATA/dup/archivelog/2010_06_06/thread_1_seq_114.316.720986321 RECID=2 STAMP=720986321
channel clone_default: deleting archived log(s)
archived log file name=+DATA/dup/archivelog/2010_06_06/thread_2_seq_83.271.720986321 RECID=1 STAMP=720986321
media recovery complete, elapsed time: 00:00:05
Finished recover at 06-JUN-10

contents of Memory Script:
{
shutdown clone immediate;
startup clone nomount;
sql clone “alter system set db_name =
”DUP” comment=
”Reset to original value by RMAN” scope=spfile”;
sql clone “alter system reset db_unique_name scope=spfile”;
shutdown clone immediate;
startup clone nomount;
}
executing Memory Script

database dismounted
Oracle instance shut down

connected to auxiliary database (not started)
Oracle instance started

Total System Global Area 849530880 bytes

Fixed Size 1339824 bytes
Variable Size 503320144 bytes
Database Buffers 339738624 bytes
Redo Buffers 5132288 bytes

sql statement: alter system set db_name = ”DUP” comment= ”Reset to original value by RMAN” scope=spfile

sql statement: alter system reset db_unique_name scope=spfile

Oracle instance shut down

connected to auxiliary database (not started)
Oracle instance started

Total System Global Area 849530880 bytes

Fixed Size 1339824 bytes
Variable Size 503320144 bytes
Database Buffers 339738624 bytes
Redo Buffers 5132288 bytes
sql statement: CREATE CONTROLFILE REUSE SET DATABASE “DUP” RESETLOGS ARCHIVELOG
MAXLOGFILES 192
MAXLOGMEMBERS 3
MAXDATAFILES 1024
MAXINSTANCES 32
MAXLOGHISTORY 292
LOGFILE
GROUP 1 SIZE 50 M ,
GROUP 2 SIZE 50 M
DATAFILE
‘+DATA/dup/datafile/system.323.720985927′
CHARACTER SET AL32UTF8

sql statement: ALTER DATABASE ADD LOGFILE

INSTANCE ‘i2′
GROUP 3 SIZE 50 M ,
GROUP 4 SIZE 50 M

contents of Memory Script:
{
set newname for clone tempfile 1 to new;
switch clone tempfile all;
catalog clone datafilecopy “+DATA/dup/datafile/sysaux.321.720985917″,
“+DATA/dup/datafile/undotbs1.319.720985933″,
“+DATA/dup/datafile/users.324.720985937″,
“+DATA/dup/datafile/example.320.720985935″,
“+DATA/dup/datafile/undotbs2.322.720985935″;
switch clone datafile all;
}
executing Memory Script

executing command: SET NEWNAME

renamed tempfile 1 to +DATA in control file

cataloged datafile copy
datafile copy file name=+DATA/dup/datafile/sysaux.321.720985917 RECID=1 STAMP=720986834
cataloged datafile copy
datafile copy file name=+DATA/dup/datafile/undotbs1.319.720985933 RECID=2 STAMP=720986836
cataloged datafile copy
datafile copy file name=+DATA/dup/datafile/users.324.720985937 RECID=3 STAMP=720986838
cataloged datafile copy
datafile copy file name=+DATA/dup/datafile/example.320.720985935 RECID=4 STAMP=720986840
cataloged datafile copy
datafile copy file name=+DATA/dup/datafile/undotbs2.322.720985935 RECID=5 STAMP=720986840

datafile 2 switched to datafile copy
input datafile copy RECID=1 STAMP=720986834 file name=+DATA/dup/datafile/sysaux.321.720985917
datafile 3 switched to datafile copy
input datafile copy RECID=2 STAMP=720986836 file name=+DATA/dup/datafile/undotbs1.319.720985933
datafile 4 switched to datafile copy
input datafile copy RECID=3 STAMP=720986838 file name=+DATA/dup/datafile/users.324.720985937
datafile 5 switched to datafile copy
input datafile copy RECID=4 STAMP=720986840 file name=+DATA/dup/datafile/example.320.720985935
datafile 6 switched to datafile copy
input datafile copy RECID=5 STAMP=720986840 file name=+DATA/dup/datafile/undotbs2.322.720985935

contents of Memory Script:
{
Alter clone database open resetlogs;
}
executing Memory Script

database opened
Finished Duplicate Db at 06-JUN-10

RMAN>

Step 7: Convert this into RAC from single instance

a) Shutdown the database.
b) Rename the spfile
c) From the init file, uncomment the RAC related parameters and rename the init file as initDUP1.ora
d) Copy this init file to node 2 under ORACLE_HOME/dbs. Save it as initDUP2.ora. Edit this file and change all DUP1 entries with DUP2 entries.

e) Update the cluster entry by adding information of the database and instance
srvctl add database -d DUP -o /u01/app/oracle/product/11.2.0/db_1
srvctl add instance -d DUP -i DUP1 -n rac1
srvctl add instance -d DUP -i DUP2 -n rac2

f) Start the cluster database DUP clone of RAC
srvctl start database -d DUP
Create spfile from pfile and restart the database.

Following steps cover all you need to do to reset SYSMAN password

1. Connect to the repository database and change the SYSMAN password in the database

SQL> alter user sysman identified by <new password> ;

2. Shutdown OMS and the agent

[oracle@localhost log]$ $OMS_HOME/opmn/bin/opmnctl stopall

3. Shutdown agent

[oracle@localhost log]$ $AGENT_HOME/bin/emctl stop agent

4. Change emoms.properties

Take a backup of this file and edit following section as shown below

eml.mntr.emdRepPwd=new_password
oracle.sysman.eml.mntr.emdRepPwdEncrypted=FALSE

5. Change targets.xml

Take a backup of this file and change the following section as follows:

<Property NAME=”password” ”newpassword” ENCRYPTED ”FALSE”/>

6. Start agent and OMS

Here we’ll perform some common maintenance tasks.

1. Adding a directory

Once a Diskgroup is created, a hierachy can be defined using directories. This is a logical way of arranging files of different categories. For example, if there is one ASM instance on a node having DATA_1 diskgroup which can house datafiles from multiple databases, then we can create directories under this Diskgroup for every database and keep the files of one database under its own directory

SQL> select name from v$asm_diskgroup ;

NAME
——————————
DATA_1

SQL> alter diskgroup data_1
2  add directory ‘+DATA_1/emrep’ ;

Diskgroup altered.

SQL> select * from v$asm_alias ;

NAME                                             GROUP_NUMBER FILE_NUMBER
———————————————— ———— ———–
FILE_INCARNATION ALIAS_INDEX ALIAS_INCARNATION PARENT_INDEX REFERENCE_INDEX A S
—————- ———– —————– ———— ————— – -
emrep                                                       1  4294967295
4294967295           0                 1     16777216        16777269 Y N

This blog covers creation of an ASM instance manually on Linux RAW Devices

Adding disk for ASM

First I added a new 500g disk on my lab server with intention of using ASM

After rebooting the server with the new disk, I created four partitions of 125G each as follows:

fdisk /dev/sdb

used n for new partition and ending cylinder was specified as +125000M to create each partition as 125G

fdisk -l showed that four disks were created as below:

/dev/sdb1

/dev/sdb2

/dev/sdb3

/dev/sdb4

Mounting RAW device

/dev/hdb1 was configured as raw device.

An entry was added in /etc/sysconfig/rawdevices file as follows:

/dev/raw/raw1 /dev/sdb1

Then rawdevice service was started

service rawdevices start

[root@localhost sysconfig]# raw -qa
/dev/raw/raw1:  bound to major 8, minor 17

Change theowner:group to oracle:dba

[root@localhost sysconfig]# ls -l /dev/raw/raw[1234]
crw-rw—-  1 oracle dba 162, 1 Jul  9 20:01 /dev/raw/raw1
[root@localhost sysconfig]#

Creation of ASM instance

ASM instance can be created using dbca utility or manually. The following example shows manual creation of the instance.

1. Create the parameter file

This is much like creation of init file for a regular instance. Specifying instance_type=asm is mandatory here.

instance_type=asm
large_pool_size=40M
background_dump_dest=/u00/app/oracle/admin/+ASM/bdump
core_dump_dest=/u00/app/oracle/admin/+ASM/cdumpuser_dump_dest=/u00/app/oracle/admin/+ASM/udump

asm_diskstring=/dev/raw/*

asm_diskstring specifies the location where disks are located. If there are multiple paths, they can be specified by seperated by comma. Wild card specification is allowed here.

2. Start the Instance

SQL>startup pfile=’/u00/app/oracle/admin/+ASM/pfile/init.ora’

Instance will be mounted but not opened gviving error on DISKGROUP not found

spfile needs be created and instance is mounted again using the spfile

SQL>create spfile from pfile=’/u00/app/oracle/admin/+ASM/pfile/init.ora’

SQL>shutdown immediate ;

SQL>startup ;

3. Creation of DISKGROUPS

DISKGROUP can be created with NORMAL (Double Mirroring), HIGH(Triple Mirroring) or External (none)

Here we create a diskgroup with External redundancy. FAILGROUP controller1 can be specified if FAILGROUPs are created. This way anything written to disks on one FAILGROUP will be written to the disks of the another FAILGROUP as well. Each FAILGROUP is connected to one disk controller. In the event of failure of one disk controller, the environment can still be availbal without any data loss. The exercise carried out here did not use any FAILGROUP though

SQL> create diskgroup DATA_1 external redundancy
2  DISK ‘/dev/raw/raw1′ ;

Diskgroup created.

SQL> select name, total_mb, free_mb from v$asm_diskgroup ;

NAME                             TOTAL_MB    FREE_MB
—————————— ———- ———-
DATA_1                             119216     119165

1. Manual Upgrade
2. Upgrade using Upgrade Assistant
3. Using Oracle Export/Data Pump

Manual upgrade gives complete control to the DBA, however, there are chance of human errors.

This blog covers the manual upgrade process from Oracle 10.2.0.1 to 11.1.0.6.

Planning for Upgrade

1.Take a good backup

2. Follow the Metalink Upgrade Document

Upgrade Pre-Work

1. Run utlu111i.sql from the ORACLE_HOME/rdbms/admin folder of the 11i software on the existing 10g databases

Following shows output of this script. It suggests if any sizing change, init/spfile parameter incompatibility,dictionary stats requirement etc.

**********************************************************************

– No update parameter changes are required.

.

**********************************************************************

Renamed Parameters: [Update Oracle Database 11.1 init.ora or spfile]

**********************************************************************

– No renamed parameters found. No changes are required.

.

**********************************************************************

Obsolete/Deprecated Parameters: [Update Oracle Database 11.1 init.ora or spfile]**********************************************************************

–> “background_dump_dest” replaced by “diagnostic_dest”

–> “user_dump_dest” replaced by “diagnostic_dest”

–> “core_dump_dest” replaced by “diagnostic_dest”

.

**********************************************************************

Components: [The following database components will be upgraded or installed]

**********************************************************************

–> Oracle Catalog Views [upgrade] VALID

–> Oracle Packages and Types [upgrade] VALID

–> Oracle Workspace Manager [upgrade] VALID

.

**********************************************************************

Miscellaneous Warnings

**********************************************************************

WARNING: –> Database is using an old timezone file version.

…. Patch the 10.2.0.1.0 database to timezone file version 4

…. BEFORE upgrading the database. Re-run utlu111i.sql after

…. patching the database to record the new timezone file version.

WARNING: –> Database contains stale optimizer statistics.

…. Refer to the 11g Upgrade Guide for instructions to update

…. statistics prior to upgrading the database.

…. Component Schemas with stale statistics:

…. SYS

WARNING: –> Database contains INVALID objects prior to upgrade.

…. USER PUBLIC has 58 INVALID objects.

…. USER SYSMAN has 37 INVALID objects.

…. USER SYS has 26 INVALID objects.

.

PL/SQL procedure successfully completed.

Imagine a table, partitioned on a date column with granularity on a daily basis. So, for every year, 365 such partitions need to be opened to accommodate all transactions for that year. I wish Oracle was smart enough to understand my needs and does that for me. With Oracle 11g, it can do exactly that.

Lets look at the example below to see how this can be done in Oracle 11g:

Here we create an old fashioned range partition table

SQL> create table test_interval_part
(id varchar2(5),
testdate date,
testdesc varchar2(2000),
num number)
partition by range (testdate)
(
partition month_0109 values less than(to_date(’01-FEB-2009′,’DD-MON-YYYY’)),
partition month_0209 values less than(to_date(’01-MAR-2009′,’DD-MON-YYYY’)),
partition month_0309 values less than(to_date(’01-APR-2009′,’DD-MON-YYYY’))
)    ;

Table created.

This will allow us to add data older than April ’09

SQL> insert into test_interval_part
2  values(‘abcde’,’01-JAN-2009′, ‘TEST DATA’,1) ;

1 row created.

SQL> insert into test_interval_part
2  values(‘abcdf’,’01-FEB-2009′,’MORE TEST DATA’,2);

1 row created.

SQL> insert into test_interval_part
2  values(‘abcdf’,’01-MAR-2009′,’SOME MORE DATA’,3) ;

1 row created.

SQL> COMMIT;

Commit complete.

As we have partitions created up to March ’09 only, data newer than March ’09 will not be accepted.

SQL> insert into test_interval_part
2   values(‘abcdg’,’01-APR-2009′,’TEST’,4) ;
insert into test_interval_part
*
ERROR at line 1:
ORA-14400: inserted partition key does not map to any partition

Now, lets see how interval partitioning introduced in Oracle 11g handles it.

An existing partitioned table can be converted to interval partitioning by a simple command without touching the data in it.

SQL> alter table test_interval_part
2  set interval(NUMTOYMINTERVAL(1,’MONTH’)) ;

Table altered.

NUMTOYMINTERVAL function is used for year-to-month literals. For day-to-second literals, NUMTODSINTERVAL is used.

Now we try to insert the same row again which failed earlier due to the lack of appropriate partition.

SQL> insert into test_interval_part
2  values(‘abcdg’,’01-APR-2009′,’TEST’,4) ;

1 row created.

SQL> commit;

SQL>  select table_name, partition_name from user_tab_partitions ;

TABLE_NAME                     PARTITION_NAME
—————————— ——————————
TEST_INTERVAL_PART             MONTH_0109
TEST_INTERVAL_PART             MONTH_0209
TEST_INTERVAL_PART             MONTH_0309
TEST_INTERVAL_PART             SYS_P41

What I don’t like here is Oracle creates a partition by a system given name. However, the good part is that you don’t have to wake up in the middle of night to add partition for some unexpected future dated data. Oracle opens a partition of the desired size for you and you can later change the name as per your convention.

SQL> alter table TEST_INTERVAL_PART
2  rename partition SYS_P41 to MONTH_0409 ;

Table altered.