1、檢視資料庫是否歸檔

[[email protected] ~]$sqlplus / as sysdba

SQL> archive log list

2、實驗過程

2.1建立測試使用者

CREATE TABLESPACE qqqDATAFILE'/u01/app/oradata/yyy/qqq01.dbf' SIZE 50M

create user qqq identified by qqq defaultTABLESPACE qqq;

grant resource,connect,dba to qqq;

conn qqq/qqq

create table tb_SalGra(

pk_SalGra_ID number primary key,

losal number,

hisal number

);

insert into tb_SalGra values (1,700,1200);

insert into tb_SalGra values (2,1201,1400);

2.2在歸檔下對資料庫做RMAN全備

[[email protected] scripts]$ ./rman_bak_level0.sh

2.3刪除tb_SalGra表及qqq使用者

檢視當前時間

SQL> select to_char(sysdate,'yyyy-mm-ddhh24:mi:ss') time from dual;

TIME

-------------------

2018-03-26 08:54:20

連線qqq使用者，刪除qqq使用者的tb_SalGra表

SQL> conn qqq/qqq

Connected.

SQL> drop table tb_SalGra;

Table dropped.

檢視當前時間

SQL> select to_char(sysdate,'yyyy-mm-ddhh24:mi:ss') time from dual;

TIME

-------------------

2018-03-26 08:57:20

連線sys使用者，刪除掉qqq使用者

SQL> conn / as sysdba

Connected.

SQL> drop user qqq cascade;

User dropped.

檢視使用者是否存在select username from all_users;

2.4基於時間點的恢復

在用RMAN恢復資料庫之前，我們先檢視一下資料庫的incarnation資訊

[[email protected] ~]$rman target /

using targetdatabase control file instead of recovery catalog

List of DatabaseIncarnations

DB Key  Inc Key DB Name  DB ID            STATUS  Reset SCN Reset Time

------- --------------- ---------------- --- ---------- ----------

1       1      YYY      2086322433       PARENT 1          15-AUG-09

2       2      YYY      2086322433       CURRENT 945184     01-MAR-18

此時資料庫的狀態是DB Key=2

關庫、起庫到MOUNT狀態，連線到RMAN，做基於時間點的不完全恢復，恢復到刪除qqq使用者時間點之前（2018-03-26 08:57:20）的狀態，即恢復qqq使用者但不恢復tb_SalGra表。

SQL> shutdownimmediate;

[[email protected] ~]$ rman target /

RMAN> startupmount;          

RMAN> run{

2> allocatechannel d1 device type disk;

3> set until time"to_date('2018-03-26 08:57:20','yyyy-mm-ddhh24:mi:ss')";

4> restoredatabase;

5> recoverdatabase;}

using target database control file instead ofrecovery catalog

allocated channel: d1

channel d1: SID=1145 device type=DISK

executing command: SET until clause

Starting restore at 26-MAR-18

channel d1: starting datafile backup set restore

channel d1: specifying datafile(s) to restorefrom backup set

channel d1: restoring datafile 00001 to/u01/app/oradata/yyy/system01.dbf

channel d1: restoring datafile 00002 to/u01/app/oradata/yyy/sysaux01.dbf

channel d1: restoring datafile 00003 to/u01/app/oradata/yyy/undotbs01.dbf

channel d1: restoring datafile 00004 to/u01/app/oradata/yyy/users01.dbf

channel d1: restoring datafile 00005 to/u01/app/oradata/yyy/test_01.dbf

channel d1: restoring datafile 00006 to/u01/app/oradata/yyy/test02.dbf

channel d1: restoring datafile 00007 to/u01/app/oradata/yyy/test3_23.dbf

channel d1: restoring datafile 00008 to/u01/app/oradata/yyy/qqq01.dbf

channel d1: reading from backup piece/file/backup/rman/db0_YYY_20180326_4csuo4pk_1_1.bkp

channel d1: piecehandle=/file/backup/rman/db0_YYY_20180326_4csuo4pk_1_1.bkptag=TAG20180326T085204

channel d1: restored backup piece 1

channel d1: restore complete, elapsed time:00:04:24

Finished restore at 26-MAR-18

Starting recover at 26-MAR-18

starting media recovery

media recovery complete, elapsed time: 00:00:02

Finished recover at 26-MAR-18

released channel: d1

RMAN> alter databaseopen resetlogs;

database opened

連線到資料庫，檢視qqq使用者是否存在，以及表是否恢復

select username from all_users;

可見恢復qqq使用者成功，tb_SalGra表尚未恢復

2.5恢復tb_SalGra表：

關庫、起庫到MOUNT狀態，在剛成功恢復資料庫的基礎上，使用RMAN做基於時間點的不完全恢復，恢復到刪除qqq使用者tb_SalGra表之前的時間點（2018-03-2608:54:20）的狀態。

sql> shutdown immediate;

RMAN> startup mount;

RMAN> list incarnation;  檢視一下資料庫的incarnation資訊

RMAN>RESET DATABASE TO INCARNATION 2;

RMAN> run{

2> allocate channel d1 device type disk;

3> set until time "to_date('2018-03-2608:54:20','yyyy-mm-dd hh24:mi:ss')";

4> restore database;

5> recover database;}

RMAN> alter databaseopen resetlogs;

database opened

連線資料庫看看tb_SalGra表是否恢復！