Sunday, August 3, 2014

Snapshot Standby Database (READ WRITE mode)

Oracle provides an unique feature where the physical standby database can be opened in READ WRITE mode to perform update able transactions. Quite often we have the standby database opened in READ Only mode for reporting purposes and optionally have the active dataguard implemented, but a snapshot standby database can be used to perform both READ and WRITE activities. Most importantly, a snapshot standby database keeps receiving the redo data from the primary database but does not apply them. These redo data received from the primary database would be applied only when the snapshot standby database is converted back to the Physical standby mode. There by the snapshot standby database provides data protection on primary database.

A snapshot standby database will allow you to make use of the data available on the physical standby database (which is a mirrored copy of the primary database). This allows the users to test the application on a standby database which has the primary data before implementing it in the Real production environment. When a physical standby database is converted to a snapshot standby database, a guaranteed restore point is automatically created. Once when the updateable transactions are completed for testing purposes on the snapshot standby database and when you are converting back the snapshot standby to physical standby, oracle flashbacks to the restore point that was created earlier which means that the transactions that were made when the standby database was opened in READ WRITE mode will be flushed out.

The only requirement to have the snapshot standby is that FRA (Flash Recovery Area) must be configured on physical standby database. It is not necessary to have flashback enabled. Below are the steps on how to convert a physical standby database to a snapshot standby database and viceversa.

Oracle Database version: 11.2.0.3 Enterprise Edition
Primary database: DCPROC

Details with respect to the primary database:
      
SQL> select status,instance_name,database_role,open_mode from v$database,v$Instance;

STATUS       INSTANCE_NAME    DATABASE_ROLE    OPEN_MODE
------------ ---------------- ---------------- --------------------
OPEN         srprim           PRIMARY          READ WRITE

SQL> select thread#,max(sequence#) from v$archived_log group by thread#;

THREAD# MAX(SEQUENCE#)
------- --------------
 1                 206

Standby database Details:

Oracle database version: 11.2.0.3 Enterprise Edition
Standby database name: DRPROC

Details with respect to the physical standby database:

SQL> select status,instance_name,database_role,open_mode from v$database,v$Instance;

STATUS       INSTANCE_NAME    DATABASE_ROLE    OPEN_MODE
------------ ---------------- ---------------- ----------------
OPEN         DRPROC           PHYSICAL STANDBY READ ONLY WITH APPLY

SQL> select thread#,max(sequence#) from v$archived_log where applied='YES' group by thread#;

THREAD# MAX(SEQUENCE#)
------- --------------
1                  206

SQL> select flashback_on from v$database;

FLASHBACK_ON
------------------
NO

You can observe that the standby database is in sync with the primary database. Below outcome shows that the Flash Recovery Area is configured on the physical standby database.
      
SQL> show parameter db_recovery_file_dest

NAME                         TYPE         VALUE
---------------------------  -----------  -------------
db_recovery_file_dest        string       +FRA_NEW
db_recovery_file_dest_size   big integer  4122M

Step 1: Cancel the Managed Recovery Process (MRP) on the physical standby database, shut it down and place it in Mount mode.

SQL> alter database recover managed standby database cancel;

Database altered.

SQL> shut immediate
Database closed.
Database dismounted.
ORACLE instance shut down.

SQL> startup mount
ORA-32004: obsolete or deprecated parameter(s) specified for RDBMS instance
ORACLE instance started.
Total System Global Area 1269366784 bytes
Fixed Size                  2227984 bytes
Variable Size             805306608 bytes
Database Buffers          452984832 bytes
Redo Buffers                8847360 bytes
Database mounted.

Step 2: Once the standby database is mounted, convert the Physical standby database to snapshot standby database.

SQL> alter database convert to snapshot standby;

Database altered.

Step 3: You can now open the snapshot standby database and check its mode.
      
SQL> alter database open;

Database altered.

SQL> select status,instance_name,database_role,open_mode from v$database,v$Instance;

STATUS      INSTANCE_NAME   DATABASE_ROLE    OPEN_MODE
----------- --------------- ---------------- ------------------
OPEN        DRPROC            SNAPSHOT STANDBY READ WRITE

Small Test on the snapshot standby database.
1. Create a user called “SNAPTEST”
2. Create a table called “TEST” whose owner is “SNAPTEST” and insert some records in it. You can also update some of the records as well.
      
SQL> create user snaptest identified by oracle;

User created.

SQL> grant connect,resource to snaptest;

Grant succeeded.

SQL> conn snaptest/oracle@srps

Connected.

SQL> create table test(code number, name char(20));

Table created.

SQL> insert into test values (100,'ARUN');

1 row created.

SQL> insert into test values(200,'SHIVU');

1 row created.

SQL> commit;

Commit complete.

SQL> select * from test;

CODE       NAME
---------- --------------------
100        ARUN
200        SHIVU

SQL> update snaptest.test set code=500 where name='ARUN';

1 row updated.

SQL> commit;

Commit complete.

SQL> select * from snaptest.test;

CODE       NAME
---------- --------------------
500        ARUN
200        SHIVU

In the mean time, you can also see that the redo data from the primary database is received by the snapshot standby database but would not be applied.

On primary database the latest sequence generated is 208 and that on the standby database, the RFS process is idle for sequence 209.

Primary:
      
SQL> select thread#,max(sequence#) from v$archived_log group by thread#;

THREAD#    MAX(SEQUENCE#)
---------- --------------
 1         208

Standby:
      
SQL> select process,status,sequence# from v$managed_standby;

PROCESS   STATUS        SEQUENCE#
--------- ------------ ----------
ARCH      CLOSING               1
ARCH      CONNECTED             0
ARCH      CONNECTED             0
ARCH      CONNECTED             0
RFS       IDLE                  0
RFS       IDLE                209
RFS       IDLE                  0

7 rows selected.

Steps on converting back a snapshot standby database to physical standby database.
Step 1: Shut down the snapshot standby database and open it in Mount mode.

SQL> shut immediate
Database closed.
Database dismounted.
ORACLE instance shut down.

SQL> startup mount
ORA-32004: obsolete or deprecated parameter(s) specified for RDBMS instance
ORACLE instance started.

Total System Global Area 1269366784 bytes
Fixed Size                  2227984 bytes
Variable Size             805306608 bytes
Database Buffers          452984832 bytes
Redo Buffers                8847360 bytes
Database mounted.

Step 2: Convert the snapshot standby database to physical standby database.
      
SQL> alter database convert to physical standby;

Database altered.

Step 3: Once done, bounce the physical standby database and start the Managed Recovery Process (MRP) on it.
      
SQL> shut immediate
ORA-01507: database not mounted

ORACLE instance shut down.

SQL> startup
ORA-32004: obsolete or deprecated parameter(s) specified for RDBMS instance
ORACLE instance started.

Total System Global Area 1269366784 bytes
Fixed Size                  2227984 bytes
Variable Size             805306608 bytes
Database Buffers          452984832 bytes
Redo Buffers                8847360 bytes
Database mounted.
Database opened.

SQL> select status,instance_name,database_role,open_mode from v$database,v$Instance;

STATUS       INSTANCE_NAME   DATABASE_ROLE     OPEN_MODE
------------ --------------  ----------------  ----------------
OPEN         DRPROC            PHYSICAL STANDBY  READ ONLY

SQL> alter database recover managed standby database disconnectfrom session;

Database altered.

SQL> select process,status,sequence# from v$managed_standby;

PROCESS   STATUS        SEQUENCE#
--------- ------------ ----------
ARCH      CONNECTED             0
ARCH      CONNECTED             0
ARCH      CONNECTED             0
ARCH      CONNECTED             0
RFS       IDLE                  0
RFS       IDLE                  0
RFS       IDLE                  0
MRP0      WAIT_FOR_LOG        213

8 rows selected.

Crosscheck whether the physical standby database is in sync with the primary database.

On Primary database:
      
SQL> select thread#,max(sequence#) from v$archived_log group bythread#;

THREAD#    MAX(SEQUENCE#)
---------- --------------
1          212

On Standby database:
      
SQL> select thread#,max(sequence#) from v$archived_log where applied='YES' group by thread#;

THREAD#    MAX(SEQUENCE#)
---------- --------------
1          212

You can see below that the transactions that were carried out when the standby database is opened in READ WRITE mode are flushed out after it was converted back to physical standby database mode.
      
SQL> select * from snaptest.test;

select * from snaptest.test
*
ERROR at line 1:
ORA-00942: table or view does not exist

SQL> select username,account_status from dba_users where username='SNAPTEST';

no rows selected

No comments:

Post a Comment