Here are the easy steps-
1)
Confirming flashback enabled on both primary and
standby databases.
2) Just before stress testing – Cancel Managed recovery on standby server
and shutdown standby server.
There was a requirement
to run stress testing (provably it may run 24 hours) in production active data
guard database environment (DB version - Enterprise
Edition Release 12.1.0.2.0 – 64bit).
Stress
testing will create lots of testing data and after that they want to rollback all
testing transactions. Means they want to
go back to the state where it was before stress testing start.
So for that, I recommended
to use flashback Database over point-in-time recovery. because it is very easy ,quick and with less impact on database availability.
I suggested following two step by step Methods as per their environment.
you guys can use any of them. both worked fine for me.
you guys can use any of them. both worked fine for me.
Two methods
are below-
Method #1:
May use if Data guard’s protechtion mode is - MAXIMUM PERFORMANCE.
else use Method #2 .
else use Method #2 .
So in this
case, we can plan to shutdown standby server before stress testing and start
after flashback database on primary database.
Steps are below-
SQL>
select database_role,flashback_on from v$database;
SQL> ALTER DATABASE RECOVER MANAGED STANDBY DATABASE CANCEL;
3) Create a guaranteed restore point in primary.
SQL> CREATE
RESTORE POINT before_stress_test_pr GUARANTEE FLASHBACK DATABASE;
4) After stress testing ,
On primary –
SQL> SHUTDOWN IMMEDIATE ;
SQL>
STARTUP MOUNT ;
SQL> FLASHBACK DATABASE TO RESTORE POINT before_stress_test_pr;
SQL> ALTER DATABASE OPEN RESETLOGS ;
Checking SCN number in primary –(just for confirmation it is greater that standby by SCN number later)
SQL> SELECT CURRENT_SCN FROM V$DATABASE;
5) Start the standby server .
6) Start Managed recovery on the standby –
SQL> ALTER DATABASE RECOVER MANAGED STANDBY DATABASE DISCONNECT FROM SESSION;
7) Drop guaranteed restore point from primary.
SQL> DROP
RESTORE POINT before_stress_test_pr;
Method #2
This one can be used any situation.
1)
Confirming flashback enabled on both primary and
standby databases.
SQL>
select database_role,flashback_on from v$database;
2)
Create a guaranteed restore point in
standby ( just before stress testing
start)
CREATE
RESTORE POINT before_stress_test_st GUARANTEE FLASHBACK DATABASE;
3)
After few minutes, create a guaranteed restore
point in primary (just before stress testing start)
CREATE
RESTORE POINT before_stress_test_pr GUARANTEE FLASHBACK DATABASE;
After
stress testing ,
4)
On
primary –
SQL> SHUTDOWN IMMEDIATE ;
SQL>
STARTUP MOUNT ;
SQL> FLASHBACK DATABASE TO RESTORE POINT before_stress_test_pr;
SQL> ALTER DATABASE OPEN RESETLOGS ;
Checking SCN number in primary –(just for confirmation it is greater that standby by SCN number later)
SQL> SELECT CURRENT_SCN FROM V$DATABASE;
5) On Standby –
SQL> ALTER DATABASE RECOVER MANAGED STANDBY DATABASE CANCEL;
Checking SCN number in standby – for confirmation it is lesser than primary SCN number.
SQL> SELECT CURRENT_SCN FROM V$DATABASE;
SQL> FLASHBACK STANDBY DATABASE TO RESTORE POINT before_stress_test_st ;
6) Start Managed recovery on the standby –
SQL> ALTER DATABASE RECOVER MANAGED STANDBY DATABASE DISCONNECT FROM SESSION;
7)
Drop both guaranteed restore point.
More Usability of Flashback Database-
Typically, in following situations you can use Flashback Database.
-
A user error or corruption removes needed
data or introduces corrupted data. For example, a user or DBA might
erroneously delete or update the contents of one or more tables, drop
database objects that are still needed during an update to an
application, or run a large batch update that fails midway.
-
A database upgrade fails or an upgrade script goes awry.
-
A complete database recovery after a media
failure cannot succeed because you do not have all of the needed redo
logs or incremental backups.
No comments:
Post a Comment