闪回数据库database
3.1.3 flashback database的配置
flashback database要求数据库必须处于归档模式,且闪回之后必须使用resetlogs打开数据库。
a.查看数据库的归档模式及闪回是否启用
SQL> select log_mode,open_mode,flashback_on from v$database;
LOG_MODE OPEN_MODE FLASHBACK_ON
------------ ---------- ------------------
ARCHIVELOG READ WRITE NO
--FLASHBACK_ON为NO,则表示闪回特性尚未启用
b.查看及设置闪回目录、闪回目录空间大小等
--下面查看恢复目录及恢复目路分配的大小
--可以使用alter system set db_recovery_file_dest 来设置新路径
--可以使用alter system set db_recovery_file_dest_size来设定新的大小
SQL> show parameter db_recovery
NAME TYPE VALUE
------------------------------------ ----------- ------------------------------
db_recovery_file_dest string /u01/app/oracle/flash_recovery
_area
db_recovery_file_dest_size big integer 2G
c.设置闪回保留目标生存期
SQL> show parameter db_flashback --缺省为分钟,即小时
NAME TYPE VALUE
------------------------------------ ----------- ------------------------------
db_flashback_retention_target integer 1440
SQL> alter system set db_flashback_retention_target=30; --设定保留时间为半小时
d.在mount状态下来启用flashback,如在open状态下则出现下列错误提示
SQL> alter database flashback on;
alter database flashback on
*
ERROR at line 1:
ORA-38759: Database must be mounted by only one instance and not open.
--一致性关闭数据库后,在mount状态下设置flashback
SQL>startup mount exclusive;
SQL> select status from v$instance;
STATUS
------------
MOUNTED
SQL> alter database flashback on;
--开启闪回数据库功能
SQL> ho ps -ef | grep rvw
--可以看到新增了后台进程rvwr
oracle 3563 1 0 12:12 ? 00:00:00 ora_rvwr_orcl
--下面查看闪回区分配的大小为大约M,闪回分钟以内的数据则需要M左右的空间
--注意列oldest_flashback_time说明了允许返回的最早的时间点
SQL>select oldest_flashback_scn,old_flhbck_scn,oldest_flashback_time old_flhbck_tim,
2 retention_target rete_trgt,flashback_size/1024/1024 flhbck_siz,
3 estimated_flashback_size/1024/1024 est_flhbck_size
4 from v$flashback_database_log;
OLD_FLHBCK_SCN OLD_FLHBC RETE_TRGT FLHBCK_SIZ EST_FLHBCK_SIZE
-------------- --------- ---------- ---------- ---------------
915137 24-OCT-10 30 7.8125 11.2519531
SQL> select * from v$flashback_database_stat; --查看闪回
BEGIN_TIM END_TIME FLASHBACK_DATA DB_DATA REDO_DATA ESTIMATED_FLASHBACK_SIZE
--------- --------- -------------- ---------- ---------- ------------------------
24-OCT-10 24-OCT-10 7905280 86802432 96329728 0
SQL> select * from v$sgastat where name like 'flashback%'; --查看sga中分配的闪回空间大小
POOL NAME BYTES
------------ -------------------------- ----------
shared pool flashback generation buff 3981204
shared pool flashback_marker_cache_si 9196
SQL> ho ls -hlt $ORACLE_BASE/flash_recovery_area/ORCL/flashback --查看生成的闪回日志
total 7.9M
-rw-r----- 1 oracle oinstall 7.9M Oct 24 12:37 o1_mf_6d7dkogw_.flb。
3.2 使用flashback database闪回数据库
步骤(前提归档日志可用)
关闭数据库
启动数据库到mount状态(exclusive模式)
闪回至某个时间点,SCN或log sequence number
使用resetlogs打开数据库
1.使用sqlplus实现闪回
可以接受一个时间标记或一个系统改变号实参
sqlplus几种常用的闪回数据库方法
FLASHBACK [STANDBY] DATABASE [<database_name>] TO [BEFORE] SCN <system_change_number> --基于SCN闪回
FLASHBACK [STANDBY] DATABASE [<database_name>] TO [BEFORE] TIMESTMP <system_timestamp_value>--基于时间戳闪回
FLASHBACK [STANDBY] DATABASE [<database_name>] TO [BEFORE] RESTORE POINT <restore_point_name>--基于时点闪回
如下面的示例:
SQL> flashback database to timestamp('2018-04-10 13:04:30','yyyy-mm-dd hh24:mi:ss');
SQL> flashback database to scn 918987;
SQL> flashback database ro restore point b1_load;
a.基于时间戳闪回
SQL> select count(1) from usr1.tb1; --查询用户usr1下表tb1中的记录数
COUNT(1)
----------
404944
SQL> select to_char(sysdate,'yyyy-mm-dd hh24:mi:ss') tm from dual; --获得系统当前的时间
TM
-------------------
2010-10-24 13:04:30
SQL> drop user usr1 cascade; --删除帐户usr1,同时帐户usr1下的所有对象将被删除
SQL> conn scott/tiger; --使用scott 帐户登陆
SQL> create table tb_emp as select * from emp; --新创建表tb_emp
SQL> shutdown immediate; --关系系统
SQL> startup mount ; --重新到mount状态
SQL> flashback database to timestamp --实施闪回
2 to_timestamp('2010-10-24 13:04:30','yyyy-mm-dd hh24:mi:ss') ;
SQL> alter database open resetlogs;
SQL> select count(1) from usr1.tb1; --帐户usr1及其对象tb1被成功闪回
COUNT(1)
----------
404944
SQL> select count(1) from scott.tb_emp; --闪回成功后,在闪回点之后修改的数据全部丢失
select count(1) from scott.tb_emp
*
ERROR at line 1:
ORA-00942: table or view does not exist
b.基于SCN号闪回
SQL> select current_scn from v$database; --获得当前的SCN号
CURRENT_SCN
-----------
918987
SQL> drop table usr1.tb1; --删除用户usr1下的表tb1
SQL> alter system checkpoint; --手动执行检查点
SQL> select file#,checkpoint_change# from v$datafile;
FILE# CHECKPOINT_CHANGE#
---------- ------------------
5 921478
SQL> shutdown abort;
SQL> startup mount;
SQL> flashback database to scn 918987;
SQL> select count(1) from usr1.tb1;
COUNT(1)
----------
404944
c.基于时点闪回
SQL> create table t(id int,col varchar2(20)); --创建表t
SQL> insert into t values(1,'ABC');
SQL> insert into t values(2,'DEF');
SQL> commit;
SQL> create restore point bef_damage; --创建闪回点
SQL> insert into t values(3,'GHI');
SQL> select ora_rowscn,id,col from t; --查看表t的记录
ORA_ROWSCN ID COL
---------- ---------- --------------------
1874406 1 ABC
1874406 2 DEF
1874406 3 GHI
SQL> shutdown immediate;
SQL> startup mount exclusive;
SQL> flashback database to restore point bef_damage; --实施时点闪回
SQL> alter database open resetlogs;
SQL> select * from t; --闪回成功后,闪回点之后的数据丢失
ID COL
---------- --------------------
1 ABC
2 DEF
Congratulations @wjwyjy! You received a personal award!
You can view your badges on your Steem Board and compare to others on the Steem Ranking
Do not miss the last post from @steemitboard:
Vote for @Steemitboard as a witness to get one more award and increased upvotes!
Downvoting a post can decrease pending rewards and make it less visible. Common reasons:
Submit