Oracle Study之--基于ASM的TSPITR（基于表空间的完全恢复）

野性酷女 2022-08-07 10:41 153阅读 0赞

## Oracle Study之--基于ASM的TSPITR（基于表空间的完全恢复） ##

**系统环境：**

**操作系统：AIX5.3-08**

**数据库：  Oracle 10gR2**

## Understanding RMAN TSPITR ##

In order to use TSPITR effectively, you need to understand what problems it can solve for you, what the major elements used in TSPITR are, what RMAN does during TSPITR, and limitations on when and how it can be applied.

### RMAN TSPITR Concepts ###

Figure 8-1, "Tablespace Point-in-Time Recovery (TSPITR) Architecture" illustrates the context within which TSPITR takes place, and a general outline of the process.

Figure 8-1 Tablespace Point-in-Time Recovery (TSPITR) Architecture

[![wKiom1TIXrnRCEMWAAGW05m6DzE496.jpg][]][wKiom1TIXrnRCEMWAAGW05m6DzE496.jpg]

Description of "Figure 8-1 Tablespace Point-in-Time Recovery (TSPITR) Architecture"

The figure contains the following entities:

*  The **target instance**, containing the tablespace to be recovered
 *  The **Recovery Manager** client
 *  The **control file** and (optional) **recovery catalog**, used for the RMAN repository records of backup activity
 *  **Archived redo logs** and **backup sets** from the target database, which are the source of the reconstructed tablespace.
 *  The **auxiliary instance**, an Oracle database instance used in the recovery process to perform the actual work of recovery.

There are four other important terms related to TSPITR, which will be used in the rest of this discussion:

*  The **target time**, the point in time or SCN that the tablespace will be left at after TSPITR
 *  The **recovery set**, which consists of the datafiles containing the tablespaces to be recovered;
 *  The **auxiliary set**, which includes datafiles required for TSPITR of the recovery set which are not themselves part of the recovery set. The auxiliary set typically includes:
    
    The auxiliary instance has other files associated with it, such as a control file, parameter file, and online logs, but they are not part of the auxiliary set.
    
     *  A copy of the SYSTEM tablespace
     *  Datafiles containing rollback or undo segments from the target instance
     *  In some cases, a temporary tablespace, used during the export of database objects from the auxiliary instance
 *  The **auxiliary destination**, an optional location on disk which can be used to store any of the auxiliary set datafiles, control files and online logs of the auxiliary instance during TSPITR. Files stored here can be deleted after TSPITR is complete.

## Performing Basic RMAN TSPITR ##

Having selected your tablespaces to recover and your target time, you are now ready to perform RMAN TSPITR. You have a few different options available to you:

*  **Fully automated TSPITR**\--in which you specify an auxiliary destination and let RMAN manage all aspects of the TSPITR. This is the simplest way to perform TSPITR, and is recommended unless you specifically need more control over the location of recovery set files after TSPITR or auxiliary set files during TSPITR, or control over the channel configurations or some other aspect of your auxiliary instance.
 *  **Customized TSPITR with an automatic auxiliary instance**\--in which you base your TSPITR on the behavior of fully automated TSPITR, possibly still using an auxiliary destination, but customize one or more aspects of the behavior, such as the location of auxiliary set or recovery set files, or specifying initialization parameters or channel configurations for the auxiliary instance created and managed by RMAN.
 *  **TSPITR with your own auxiliary instance**\--in which you take responsibility for setting up, starting, stopping and cleaning up the auxiliary instance used in TSPITR, and possibly also manage the TSPITR process using some of the methods available in customized TSPITR with an automatic auxiliary instance.

**案例分析：**

**    用户在过去的时间点对emp1表做了误操作，需要通过物理备份恢复到过去的数据，本案例采用基于表空间的不完全恢复来实现对表的恢复！**

**1、案例测试环境**

SQL> show parameter name
    NAME                                 TYPE        VALUE
    ------------------------------------ ----------- ------------------------------
    db_file_name_convert                 string
    db_name                              string      zsdb
    db_unique_name                       string      zsdb
    global_names                         boolean     FALSE
    instance_name                        string      zsdb
    lock_name_space                      string
    log_file_name_convert                string
    service_names                        string      zsdb
    SQL> select name from v$datafile;
    NAME
    --------------------------------------------------------------------------------
    +DG1/zsdb/datafile/system.257.870003801
    +DG1/zsdb/datafile/undotbs1.260.870004111
    +DG1/zsdb/datafile/sysaux.258.870003999
    +DG1/zsdb/datafile/users.262.870004141
    +DG1/zsdb/datafile/zstb.259.870004085
    +DG1/zsdb/datafile/testtb.261.870004125
    6 rows selected.
    SQL> col member for a50
    SQL> select group#,member from v$logfile;
        GROUP# MEMBER
    ---------- --------------------------------------------------
             4 +DG1/zsdb/onlinelog/group_4.266.870004663
             4 +RCY1/zsdb/onlinelog/group_4.256.870004669
             5 +DG1/zsdb/onlinelog/group_5.267.870004689
             5 +RCY1/zsdb/onlinelog/group_5.257.870004693
    SQL> set linesize 120
    SQL> r
      1* select file_id,file_name,tablespace_name from dba_data_files
       FILE_ID FILE_NAME                                          TABLESPACE_NAME
    ---------- -------------------------------------------------- ------------------------------
             4 +DG1/zsdb/datafile/users.262.870004141             USERS
             3 +DG1/zsdb/datafile/sysaux.258.870003999            SYSAUX
             2 +DG1/zsdb/datafile/undotbs1.260.870004111          UNDOTBS1
             1 +DG1/zsdb/datafile/system.257.870003801            SYSTEM
             5 +DG1/zsdb/datafile/zstb.259.870004085              ZSTB
             6 +DG1/zsdb/datafile/testtb.261.870004125            TESTTB
    6 rows selected.

**2、建立测试**

SQL> select * from tab;
    TNAME                          TABTYPE  CLUSTERID
    ------------------------------ ------- ----------
    DEPT                           TABLE
    EMP                            TABLE
    BONUS                          TABLE
    SALGRADE                       TABLE
    TB_EMP                         TABLE
    TB_DEPT                        TABLE
    TB1                            TABLE
    EMP1                           TABLE
    8 rows selected.
    
    SQL> select table_name,tablespace_name from user_tables;
    TABLE_NAME                     TABLESPACE_NAME
    ------------------------------ ------------------------------
    DEPT                           USERS
    EMP                            USERS
    BONUS                          USERS
    SALGRADE                       USERS
    TB_EMP                         TESTTB
    TB_DEPT                        TESTTB
    TB1                            USERS
    EMP1                           ZSTB
    8 rows selected.
    
    10:43:58 SQL> insert into emp1 select * from emp;
    14 rows created.
    
    10:44:10 SQL> commit;
    Commit complete.
    
    10:44:12 SQL> select * from emp1;
         EMPNO ENAME      JOB              MGR HIREDATE         SAL       COMM     DEPTNO
    ---------- ---------- --------- ---------- --------- ---------- ---------- ----------
          9999 SMITH      CLERK           7902 17-DEC-80        800                    20
          7369 SMITH      CLERK           7902 17-DEC-80        800                    20
          7499 ALLEN      SALESMAN        7698 20-FEB-81       1600        300         30
          7521 WARD       SALESMAN        7698 22-FEB-81       1250        500         30
          7566 JONES      MANAGER         7839 02-APR-81       2975                    20
          7654 MARTIN     SALESMAN        7698 28-SEP-81       1250       1400         30
          7698 BLAKE      MANAGER         7839 01-MAY-81       2850                    30
          7782 CLARK      MANAGER         7839 09-JUN-81       2450                    10
          7788 SCOTT      ANALYST         7566 19-APR-87       3000                    20
          7839 KING       PRESIDENT            17-NOV-81       5000                    10
          7844 TURNER     SALESMAN        7698 08-SEP-81       1500          0         30
         EMPNO ENAME      JOB              MGR HIREDATE         SAL       COMM     DEPTNO
    ---------- ---------- --------- ---------- --------- ---------- ---------- ----------
          7876 ADAMS      CLERK           7788 23-MAY-87       1100                    20
          7900 JAMES      CLERK           7698 03-DEC-81        950                    30
          7902 FORD       ANALYST         7566 03-DEC-81       3000                    20
          7934 MILLER     CLERK           7782 23-JAN-82       1300                    10
    15 rows selected.
    
    10:44:34 SQL> update emp1 set sal=9000 ;    
    15 rows updated.
    
    10:44:47 SQL> commit;
    Commit complete.
    
    10:44:49 SQL> select * from emp1;
         EMPNO ENAME      JOB              MGR HIREDATE         SAL       COMM     DEPTNO
    ---------- ---------- --------- ---------- --------- ---------- ---------- ----------
          9999 SMITH      CLERK           7902 17-DEC-80       9000                    20
          7369 SMITH      CLERK           7902 17-DEC-80       9000                    20
          7499 ALLEN      SALESMAN        7698 20-FEB-81       9000        300         30
          7521 WARD       SALESMAN        7698 22-FEB-81       9000        500         30
          7566 JONES      MANAGER         7839 02-APR-81       9000                    20
          7654 MARTIN     SALESMAN        7698 28-SEP-81       9000       1400         30
          7698 BLAKE      MANAGER         7839 01-MAY-81       9000                    30
          7782 CLARK      MANAGER         7839 09-JUN-81       9000                    10
          7788 SCOTT      ANALYST         7566 19-APR-87       9000                    20
          7839 KING       PRESIDENT            17-NOV-81       9000                    10
          7844 TURNER     SALESMAN        7698 08-SEP-81       9000          0         30
         EMPNO ENAME      JOB              MGR HIREDATE         SAL       COMM     DEPTNO
    ---------- ---------- --------- ---------- --------- ---------- ---------- ----------
          7876 ADAMS      CLERK           7788 23-MAY-87       9000                    20
          7900 JAMES      CLERK           7698 03-DEC-81       9000                    30
          7902 FORD       ANALYST         7566 03-DEC-81       9000                    20
          7934 MILLER     CLERK           7782 23-JAN-82       9000                    10
    15 rows selected.
    
    表被误操作前的scn：
    SQL> select current_scn from v$database;
    CURRENT_SCN
    -----------
         988659
         
    10:44:54 SQL> truncate table emp1 ;
    Table truncated.
    
    10:46:28 SQL> select * from emp1;
    no rows selected
    
    10:46:33 SQL> insert into emp1 select * from emp where rownum=1;
    1 row created.
    
    10:46:45 SQL> update emp1 set empno=9999;
    1 row updated.
    
    10:46:53 SQL> commit;
    Commit complete.
    
    10:46:56 SQL> select * from emp1;
         EMPNO ENAME      JOB              MGR HIREDATE         SAL       COMM     DEPTNO
    ---------- ---------- --------- ---------- --------- ---------- ---------- ----------
          9999 SMITH      CLERK           7902 17-DEC-80        800                    20

**3、检查表空间自包含**

SQL> exec DBMS_TTS.TRANSPORT_SET_CHECK('ZSTB',TRUE);
    PL/SQL procedure successfully completed.
    SQL> select * from TRANSPORT_SET_VIOLATIONS t;
    no rows selected

**4、通过rman做TSPITR**

[10:49:12 oracle@aix217: dbs]$mkdir /home/oracle/zsdb
    [10:49:23 oracle@aix217: dbs]$cd /home/oracle/zsdb
    [10:49:30 oracle@aix217: zsdb]$rman target /
    Recovery Manager: Release 10.2.0.1.0 - Production on Wed Jan 28 10:49:35 2015
    Copyright (c) 1982, 2005, Oracle.  All rights reserved.
    connected to target database: ZSDB (DBID=3945368821)
    RMAN>  recover tablespace zstb until scn 988659
    2> ;
    Starting recover at 28-JAN-15
    using target database control file instead of recovery catalog
    allocated channel: ORA_DISK_1
    channel ORA_DISK_1: sid=125 devtype=DISK
    RMAN-05026: WARNING: presuming following set of tablespaces applies to specified point in time
    List of tablespaces expected to have UNDO segments
    tablespace SYSTEM
    tablespace UNDOTBS1
    Creating automatic instance, with SID='pgoi'
    initialization parameters used for automatic instance:
    db_name=ZSDB
    compatible=10.2.0.1.0
    db_block_size=8192
    db_files=200
    db_unique_name=tspitr_ZSDB_pgoi
    large_pool_size=1M
    shared_pool_size=110M
    #No auxiliary parameter file used
    #No auxiliary destination in use
    #Use default controlfile
    starting up automatic instance ZSDB
    Oracle instance started
    Total System Global Area     205520896 bytes
    Fixed Size                     2019680 bytes
    Variable Size                146804384 bytes
    Database Buffers              50331648 bytes
    Redo Buffers                   6365184 bytes
    Automatic instance created
    Removing automatic instance
    shutting down automatic instance 
    Oracle instance shut down
    Automatic instance removed
    RMAN-00571: ===========================================================
    RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
    RMAN-00571: ===========================================================
    RMAN-03002: failure of recover command at 01/28/2015 10:50:09
    RMAN-05002: aborting point-in-time tablespace recovery
    RMAN-05517: temporary file +DG1/zsdb/tempfile/temp.265.870004489 conflicts with file used by target database
    RMAN-05001: auxiliary filename +DG1/zsdb/datafile/undotbs1.260.870004111 conflicts with a file used by the target database
    RMAN-05001: auxiliary filename +DG1/zsdb/datafile/system.257.870003801 conflicts with a file used by the target database

\---在做TSPITR时，需要建立Auxiliary Instance，需要转储system、undotbs，应该将Auxiliary Instance的datafile与原库存储到不同的diskgroup。

**5、建立ASM diskgroup**

[11:05:43 root@aix217: /]#lsvg rootvg
    VOLUME GROUP:       rootvg                   VG IDENTIFIER:  0000c9a20000d700000001447571be5b
    VG STATE:           active                   PP SIZE:        128 megabyte(s)
    VG PERMISSION:      read/write               TOTAL PPs:      546 (69888 megabytes)
    MAX LVs:            256                      FREE PPs:       85 (10880 megabytes)
    LVs:                14                       USED PPs:       461 (59008 megabytes)
    OPEN LVs:           12                       QUORUM:         2 (Enabled)
    TOTAL PVs:          1                        VG DESCRIPTORS: 2
    STALE PVs:          0                        STALE PPs:      0
    ACTIVE PVs:         1                        AUTO ON:        yes
    MAX PPs per VG:     32512                                     
    MAX PPs per PV:     1016                     MAX PVs:        32
    LTG size (Dynamic): 256 kilobyte(s)          AUTO SYNC:      no
    HOT SPARE:          no                       BB POLICY:      relocatable 
    
    [11:06:27 root@aix217: /]#mklv -y 'lv_aux' -t 'raw' rootvg 30
    lv_aux
    
    [11:07:39 root@aix217: /]#ls -l /dev |grep aux
    brw-rw----    1 root     system       10, 15 Jan 28 11:07 lv_aux
    brw-rw----    1 oracle   dba          70,  7 Jan 14 16:26 rac_sysaux
    crw-rw----    1 root     system       10, 15 Jan 28 11:07 rlv_aux
    crw-rw----    1 oracle   dba          70,  7 Jan 14 16:26 rrac_sysaux
    
    [11:07:50 root@aix217: /]#chown oracle:dba /dev/*lv_aux
    [11:08:21 root@aix217: /]#chmod 660 /dev/*lv_aux
    [11:08:29 root@aix217: /]#ls -l /dev |grep aux
    brw-rw----    1 oracle   dba          10, 15 Jan 28 11:07 lv_aux
    crw-rw----    1 oracle   dba          10, 15 Jan 28 11:07 rlv_aux
    
    SQL> create diskgroup dg2 external redundancy
      2  disk '/dev/rlv_aux';
    Diskgroup created.
    SQL> select name ,state from v$asm_diskgroup;
    NAME                           STATE
    ------------------------------ -----------
    DG1                            MOUNTED
    RCY1                           MOUNTED
    DG2                            MOUNTED

**6、再次做TSPITR**

[11:13:58 oracle@aix217: ~]$rman target /
    Recovery Manager: Release 10.2.0.1.0 - Production on Wed Jan 28 11:14:07 2015
    Copyright (c) 1982, 2005, Oracle.  All rights reserved.
    RMAN-06900: WARNING: unable to generate V$RMAN_STATUS or V$RMAN_OUTPUT row
    RMAN-06901: WARNING: disabling update of the V$RMAN_STATUS and V$RMAN_OUTPUT rows
    ORACLE error from target database: 
    ORA-29701: unable to connect to Cluster Manager
    connected to target database: ZSDB (DBID=3945368821)
    RMAN> RUN
    2> {
    3>   SET NEWNAME FOR DATAFILE 1 to '+dg2';
    4>    SET NEWNAME FOR DATAFILE 2 to '+dg2';
    5>   RECOVER TABLESPACE zstb until scn 988659
    6>      AUXILIARY DESTINATION '+DG2';
    }
    7> 
    executing command: SET NEWNAME
    using target database control file instead of recovery catalog
    executing command: SET NEWNAME
    Starting recover at 28-JAN-15
    allocated channel: ORA_DISK_1
    channel ORA_DISK_1: sid=125 devtype=DISK
    RMAN-05026: WARNING: presuming following set of tablespaces applies to specified point in time
    List of tablespaces expected to have UNDO segments
    tablespace SYSTEM
    tablespace UNDOTBS1
    
    建立并启动辅助实例：
    Creating automatic instance, with SID='qauC'
    initialization parameters used for automatic instance:
    db_name=ZSDB
    compatible=10.2.0.1.0
    db_block_size=8192
    db_files=200
    db_unique_name=tspitr_ZSDB_qauC
    large_pool_size=1M
    shared_pool_size=110M
    #No auxiliary parameter file used
    db_create_file_dest=+DG2
    control_files=+DG2/cntrl_tspitr_ZSDB_qauC.f
    
    
    starting up automatic instance ZSDB
    Oracle instance started
    Total System Global Area     205520896 bytes
    Fixed Size                     2019680 bytes
    Variable Size                146804384 bytes
    Database Buffers              50331648 bytes
    Redo Buffers                   6365184 bytes
    Automatic instance created
    contents of Memory Script:
    {
    # set the until clause
    set until  scn 988659;
    # restore the controlfile
    restore clone controlfile;
    # mount the controlfile
    sql clone 'alter database mount clone database';
    # archive current online log for tspitr to a resent until time
    sql 'alter system archive log current';
    # avoid unnecessary autobackups for structural changes during TSPITR
    sql 'begin dbms_backup_restore.AutoBackupFlag(FALSE); end;';
    }
    executing Memory Script
    executing command: SET until clause
    Starting restore at 28-JAN-15
    allocated channel: ORA_AUX_DISK_1
    channel ORA_AUX_DISK_1: sid=33 devtype=DISK
    channel ORA_AUX_DISK_1: starting datafile backupset restore
    channel ORA_AUX_DISK_1: restoring control file
    channel ORA_AUX_DISK_1: reading from backup piece /backup/backupctl_ZSDB_ncptrifa_1_1.ctl
    channel ORA_AUX_DISK_1: restored backup piece 1
    piece handle=/backup/backupctl_ZSDB_ncptrifa_1_1.ctl tag=BKCTL
    channel ORA_AUX_DISK_1: restore complete, elapsed time: 00:00:38
    output filename=+DG2/cntrl_tspitr_zsdb_qauc.f
    Finished restore at 28-JAN-15
    sql statement: alter database mount clone database
    sql statement: alter system archive log current
    sql statement: begin dbms_backup_restore.AutoBackupFlag(FALSE); end;
    released channel: ORA_DISK_1
    released channel: ORA_AUX_DISK_1
    contents of Memory Script:
    {
    # generated tablespace point-in-time recovery script
    # set the until clause
    set until  scn 988659;
    plsql <<<-- tspitr_2
    declare
      sqlstatement       varchar2(512);
      offline_not_needed exception;
      pragma exception_init(offline_not_needed, -01539);
    begin
      sqlstatement := 'alter tablespace '||  'ZSTB' ||' offline for recover';
      krmicd.writeMsg(6162, sqlstatement);
      krmicd.execSql(sqlstatement);
    exception
      when offline_not_needed then
        null;
    end; >>>;
    # set a destination filename for restore
    set newname for datafile  1 to 
     "+DG2";
    # set a destination filename for restore
    set newname for datafile  2 to 
     "+DG2";
    # set an omf destination tempfile
    set newname for clone tempfile  2 to new;
    # set a destination filename for restore
    set newname for datafile  5 to 
     "+DG1/zsdb/datafile/zstb.259.870004085";
    # rename all tempfiles
    switch clone tempfile all;
    # restore the tablespaces in the recovery set plus the auxilliary tablespaces
    restore clone datafile  1, 2, 5;
    switch clone datafile all;
    #online the datafiles restored or flipped
    sql clone "alter database datafile  1 online";
    #online the datafiles restored or flipped
    sql clone "alter database datafile  2 online";
    #online the datafiles restored or flipped
    sql clone "alter database datafile  5 online";
    # make the controlfile point at the restored datafiles, then recover them
    recover clone database tablespace  "ZSTB", "SYSTEM", "UNDOTBS1" delete archivelog;
    alter clone database open resetlogs;
    # PLUG HERE the creation of a temporary tablespace if export fails due to lack
    # of temporary space.
    # For example in Unix these two lines would do that:
    #sql clone "create tablespace aux_tspitr_tmp
    #           datafile ''/tmp/aux_tspitr_tmp.dbf'' size 500K";
    }
    executing Memory Script
    executing command: SET until clause
    sql statement: alter tablespace ZSTB offline for recover
    executing command: SET NEWNAME
    executing command: SET NEWNAME
    executing command: SET NEWNAME
    executing command: SET NEWNAME
    renamed temporary file 2 to +DG2 in control file
    Starting restore at 28-JAN-15
    allocated channel: ORA_AUX_DISK_1
    channel ORA_AUX_DISK_1: sid=36 devtype=DISK
    channel ORA_AUX_DISK_1: starting datafile backupset restore
    channel ORA_AUX_DISK_1: specifying datafile(s) to restore from backup set
    restoring datafile 00002 to +DG2
    restoring datafile 00005 to +DG1/zsdb/datafile/zstb.259.870004085
    channel ORA_AUX_DISK_1: reading from backup piece /backup/inc0_ZSDB_mbptna6f_1_1
    channel ORA_AUX_DISK_1: restored backup piece 1
    piece handle=/backup/inc0_ZSDB_maptna6d_1_1 tag=INC0
    channel ORA_AUX_DISK_1: restore complete, elapsed time: 00:01:05
    Finished restore at 28-JAN-15
    datafile 1 switched to datafile copy
    input datafile copy recid=48 stamp=870175071 filename=+DG2/tspitr_zsdb_qauc/datafile/system.258.870175007
    datafile 2 switched to datafile copy
    input datafile copy recid=49 stamp=870175071 filename=+DG2/tspitr_zsdb_qauc/datafile/undotbs1.257.870174973
    sql statement: alter database datafile  1 online
    sql statement: alter database datafile  2 online
    sql statement: alter database datafile  5 online
    Starting recover at 28-JAN-15
    using channel ORA_AUX_DISK_1
    channel ORA_AUX_DISK_1: starting incremental datafile backupset restore
    channel ORA_AUX_DISK_1: specifying datafile(s) to restore from backup set
    destination for restore of datafile 00001: +DG2/tspitr_zsdb_qauc/datafile/system.258.870175007
    channel ORA_AUX_DISK_1: reading from backup piece /backup/inc1_ZSDB_mhptnb2j_1_1
    channel ORA_AUX_DISK_1: restored backup piece 1
    piece handle=/backup/inc1_ZSDB_mhptnb2j_1_1 tag=INC1
    channel ORA_AUX_DISK_1: restore complete, elapsed time: 00:00:01
    channel ORA_AUX_DISK_1: starting incremental datafile backupset restore
    channel ORA_AUX_DISK_1: specifying datafile(s) to restore from backup set
    destination for restore of datafile 00005: +DG1/zsdb/datafile/zstb.259.870004085
    destination for restore of datafile 00002: +DG2/tspitr_zsdb_qauc/datafile/undotbs1.257.870174973
    channel ORA_AUX_DISK_1: reading from backup piece /backup/inc1_ZSDB_miptnb2l_1_1
    channel ORA_AUX_DISK_1: restored backup piece 1
    piece handle=/backup/inc1_ZSDB_miptnb2l_1_1 tag=INC1
    channel ORA_AUX_DISK_1: restore complete, elapsed time: 00:00:02
    channel ORA_AUX_DISK_1: starting incremental datafile backupset restore
    channel ORA_AUX_DISK_1: specifying datafile(s) to restore from backup set
    destination for restore of datafile 00001: +DG2/tspitr_zsdb_qauc/datafile/system.258.870175007
    channel ORA_AUX_DISK_1: reading from backup piece /backup/inc1_ZSDB_moptnbuk_1_1
    channel ORA_AUX_DISK_1: restored backup piece 1
    piece handle=/backup/inc1_ZSDB_moptnbuk_1_1 tag=INC1
    channel ORA_AUX_DISK_1: restore complete, elapsed time: 00:00:01
    channel ORA_AUX_DISK_1: starting incremental datafile backupset restore
    channel ORA_AUX_DISK_1: specifying datafile(s) to restore from backup set
    destination for restore of datafile 00005: +DG1/zsdb/datafile/zstb.259.870004085
    destination for restore of datafile 00002: +DG2/tspitr_zsdb_qauc/datafile/undotbs1.257.870174973
    channel ORA_AUX_DISK_1: reading from backup piece /backup/inc1_ZSDB_mpptnbum_1_1
    channel ORA_AUX_DISK_1: restored backup piece 1
    piece handle=/backup/inc1_ZSDB_mpptnbum_1_1 tag=INC1
    channel ORA_AUX_DISK_1: restore complete, elapsed time: 00:00:02
    channel ORA_AUX_DISK_1: starting incremental datafile backupset restore
    channel ORA_AUX_DISK_1: specifying datafile(s) to restore from backup set
    destination for restore of datafile 00001: +DG2/tspitr_zsdb_qauc/datafile/system.258.870175007
    channel ORA_AUX_DISK_1: reading from backup piece /backup/inc1_ZSDB_mvptncqp_1_1
    channel ORA_AUX_DISK_1: restored backup piece 1
    piece handle=/backup/inc1_ZSDB_mvptncqp_1_1 tag=INC1
    channel ORA_AUX_DISK_1: restore complete, elapsed time: 00:00:02
    channel ORA_AUX_DISK_1: starting incremental datafile backupset restore
    channel ORA_AUX_DISK_1: specifying datafile(s) to restore from backup set
    destination for restore of datafile 00005: +DG1/zsdb/datafile/zstb.259.870004085
    destination for restore of datafile 00002: +DG2/tspitr_zsdb_qauc/datafile/undotbs1.257.870174973
    channel ORA_AUX_DISK_1: reading from backup piece /backup/inc1_ZSDB_n0ptncqr_1_1
    channel ORA_AUX_DISK_1: restored backup piece 1
    piece handle=/backup/inc1_ZSDB_n0ptncqr_1_1 tag=INC1
    channel ORA_AUX_DISK_1: restore complete, elapsed time: 00:00:02
    channel ORA_AUX_DISK_1: starting incremental datafile backupset restore
    channel ORA_AUX_DISK_1: specifying datafile(s) to restore from backup set
    destination for restore of datafile 00001: +DG2/tspitr_zsdb_qauc/datafile/system.258.870175007
    channel ORA_AUX_DISK_1: reading from backup piece /backup/inc1_ZSDB_n6ptridv_1_1
    channel ORA_AUX_DISK_1: restored backup piece 1
    piece handle=/backup/inc1_ZSDB_n6ptridv_1_1 tag=INC1
    channel ORA_AUX_DISK_1: restore complete, elapsed time: 00:00:02
    channel ORA_AUX_DISK_1: starting incremental datafile backupset restore
    channel ORA_AUX_DISK_1: specifying datafile(s) to restore from backup set
    destination for restore of datafile 00005: +DG1/zsdb/datafile/zstb.259.870004085
    destination for restore of datafile 00002: +DG2/tspitr_zsdb_qauc/datafile/undotbs1.257.870174973
    channel ORA_AUX_DISK_1: reading from backup piece /backup/inc1_ZSDB_n7ptrie1_1_1
    channel ORA_AUX_DISK_1: restored backup piece 1
    piece handle=/backup/inc1_ZSDB_n7ptrie1_1_1 tag=INC1
    channel ORA_AUX_DISK_1: restore complete, elapsed time: 00:00:01
    starting media recovery
    channel ORA_AUX_DISK_1: starting archive log restore to default destination
    channel ORA_AUX_DISK_1: restoring archive log
    archive log thread=1 sequence=181
    channel ORA_AUX_DISK_1: reading from backup piece /backup/arch_ZSDB_nbptriev_1_1
    channel ORA_AUX_DISK_1: restored backup piece 1
    piece handle=/backup/arch_ZSDB_nbptriev_1_1 tag=BACKCH
    channel ORA_AUX_DISK_1: restore complete, elapsed time: 00:00:01
    archive log filename=/u01/app/oracle/product/10.2.0/db_1/dbs/arch1_181_869671369.dbf thread=1 sequence=181
    channel clone_default: deleting archive log(s)
    archive log filename=/u01/app/oracle/product/10.2.0/db_1/dbs/arch1_181_869671369.dbf recid=568 stamp=870175096
    channel ORA_AUX_DISK_1: starting archive log restore to default destination
    channel ORA_AUX_DISK_1: restoring archive log
    archive log thread=1 sequence=182
    channel ORA_AUX_DISK_1: reading from backup piece /backup/arch_ZSDB_nhptrjca_1_1
    channel ORA_AUX_DISK_1: restored backup piece 1
    piece handle=/backup/arch_ZSDB_nhptrjca_1_1 tag=BACKCH
    channel ORA_AUX_DISK_1: restore complete, elapsed time: 00:00:04
    archive log filename=/u01/app/oracle/product/10.2.0/db_1/dbs/arch1_182_869671369.dbf thread=1 sequence=182
    channel clone_default: deleting archive log(s)
    archive log filename=/u01/app/oracle/product/10.2.0/db_1/dbs/arch1_182_869671369.dbf recid=569 stamp=870175100
    media recovery complete, elapsed time: 00:00:02
    Finished recover at 28-JAN-15
    database opened
    contents of Memory Script:
    {
    # export the tablespaces in the recovery set
    host 'exp userid =\"/@$DESCRIPTION=\(ADDRESS=\(PROTOCOL=beq$$PROGRAM=/u01/app/oracle/product/10.2.0/db_1/bin/oracle$$ARGV0=oracleqauC$$ARGS=^'\(DESCRIPTION=\(LOCAL=YES$$ADDRESS=\(PROTOCOL=beq$\)\)^'\)$ENVS=^'ORACLE_SID=qauC^'$\)$CONNECT_DATA=\(SID=qauC$\)\) as sysdba\" point_in_time_recover=y tablespaces=
     ZSTB file=
    tspitr_a.dmp';
    # shutdown clone before import
    shutdown clone immediate
    # import the tablespaces in the recovery set
    host 'imp userid =\"/@ as sysdba\" point_in_time_recover=y file=
    tspitr_a.dmp';
    # online/offline the tablespace imported
    sql "alter tablespace  ZSTB online";
    sql "alter tablespace  ZSTB offline";
    # enable autobackups in case user does open resetlogs from RMAN after TSPITR
    sql 'begin dbms_backup_restore.AutoBackupFlag(TRUE); end;';
    }
    executing Memory Script
    
    从Auxiliary Instance导出恢复后的数据:
    Export: Release 10.2.0.1.0 - Production on Wed Jan 28 11:18:41 2015
    Copyright (c) 1982, 2005, Oracle.  All rights reserved.
    Connected to: Oracle Database 10g Enterprise Edition Release 10.2.0.1.0 - 64bit Production
    With the Partitioning, Real Application Clusters, OLAP and Data Mining options
    Export done in US7ASCII character set and AL16UTF16 NCHAR character set
    server uses ZHS16GBK character set (possible charset conversion)
    Note: table data (rows) will not be exported
    About to export Tablespace Point-in-time Recovery objects...
    For tablespace ZSTB ...
    . exporting cluster definitions
    . exporting table definitions
    . . exporting table                           EMP1
    EXP-00091: Exporting questionable statistics.
    . . exporting table                           TEST
    . exporting referential integrity constraints
    . exporting triggers
    . end point-in-time recovery
    Export terminated successfully with warnings.
    host command complete
    database closed
    database dismounted
    Oracle instance shut down
    Import: Release 10.2.0.1.0 - Production on Wed Jan 28 11:19:25 2015
    Copyright (c) 1982, 2005, Oracle.  All rights reserved.
    Connected to: Oracle Database 10g Enterprise Edition Release 10.2.0.1.0 - 64bit Production
    With the Partitioning, Real Application Clusters, OLAP and Data Mining options
    Export file created by EXPORT:V10.02.01 via conventional path
    About to import Tablespace Point-in-time Recovery objects...
    
    导入恢复后的数据到主库：
    import done in US7ASCII character set and AL16UTF16 NCHAR character set
    import server uses ZHS16GBK character set (possible charset conversion)
    . importing SYS's objects into SYS
    . importing SCOTT's objects into SCOTT
    . . importing table                         "EMP1"
    . importing TEST's objects into TEST
    . . importing table                         "TEST"
    . importing SYS's objects into SYS
    Import terminated successfully without warnings.
    host command complete
    sql statement: alter tablespace  ZSTB online
    sql statement: alter tablespace  ZSTB offline
    sql statement: begin dbms_backup_restore.AutoBackupFlag(TRUE); end;
    Removing automatic instance
    Automatic instance removed
    auxiliary instance file +DG2/cntrl_tspitr_zsdb_qauc.f deleted
    Removing automatic instance
    Automatic instance removed
    RMAN-00571: ===========================================================
    RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
    RMAN-00571: ===========================================================
    RMAN-03002: failure of recover command at 01/28/2015 11:19:50
    ORA-15045: ASM file name '+DG2' is not in reference form

错误信息：
        在recovery完成后，会删除Auxiliary Instance的文件，因为文件放在ASM diskgroup下，故删除失败，可以考虑将Auxiliary Instance文件放在文件系统下。
        [11:32:21 oracle@aix217: ~]$asmcmd
    ASMCMD> ls
    DG1/
    DG2/
    RCY1/
    ASMCMD> cd dg2
    ASMCMD> ls
    TSPITR_ZSDB_QAUC/
    ASMCMD> cd TSPITR_ZSDB_QAUC/
    ASMCMD> ls
    DATAFILE/
    ONLINELOG/
    TEMPFILE/
    ASMCMD>
    
    恢复脚本：        
    RUN
     {
       SET NEWNAME FOR DATAFILE 1 to '+dg2';
       SET NEWNAME FOR DATAFILE 2 to '+dg2';
       RECOVER TABLESPACE zstb until scn 988659
       AUXILIARY DESTINATION '/home/oracle/zsdb';
    }

**7、验证数据恢复**

SQL> conn scott/tiger
    Connected.
    SQL> select * from emp1;
    select * from emp1
                  *
    ERROR at line 1:
    ORA-00376: file 5 cannot be read at this time
    ORA-01110: data file 5: '+DG1/zsdb/datafile/zstb.259.870004085'
    SQL> conn /as sysdba
    Connected.
    SQL> alter database datafile 5 online;
    Database altered.
    SQL> conn scott/tiger
    Connected.
    SQL> select * from emp1;
    select * from emp1
                  *
    ERROR at line 1:
    ORA-00376: file 5 cannot be read at this time
    ORA-01110: data file 5: '+DG1/zsdb/datafile/zstb.259.870004085'
    
    SQL> alter tablespace zstb online;
    Tablespace altered.
    SQL> set linesize 120
    SQL> r
      1* select * from scott.emp1
         EMPNO ENAME      JOB              MGR HIREDATE         SAL       COMM     DEPTNO
    ---------- ---------- --------- ---------- --------- ---------- ---------- ----------
          9999 SMITH      CLERK           7902 17-DEC-80       9000                    20
          7369 SMITH      CLERK           7902 17-DEC-80       9000                    20
          7499 ALLEN      SALESMAN        7698 20-FEB-81       9000        300         30
          7521 WARD       SALESMAN        7698 22-FEB-81       9000        500         30
          7566 JONES      MANAGER         7839 02-APR-81       9000                    20
          7654 MARTIN     SALESMAN        7698 28-SEP-81       9000       1400         30
          7698 BLAKE      MANAGER         7839 01-MAY-81       9000                    30
          7782 CLARK      MANAGER         7839 09-JUN-81       9000                    10
          7788 SCOTT      ANALYST         7566 19-APR-87       9000                    20
          7839 KING       PRESIDENT            17-NOV-81       9000                    10
          7844 TURNER     SALESMAN        7698 08-SEP-81       9000          0         30
         EMPNO ENAME      JOB              MGR HIREDATE         SAL       COMM     DEPTNO
    ---------- ---------- --------- ---------- --------- ---------- ---------- ----------
          7876 ADAMS      CLERK           7788 23-MAY-87       9000                    20
          7900 JAMES      CLERK           7698 03-DEC-81       9000                    30
          7902 FORD       ANALYST         7566 03-DEC-81       9000                    20
          7934 MILLER     CLERK           7782 23-JAN-82       9000                    10
    15 rows selected.

**@至此，emp1表中被误删除的数据被恢复完成！**

**通过RMAN做基于表空间的不完全恢复（TSPTIR),可以在不关闭数据库的情况下快速恢复因为用户误操作而丢失的数据；相比传统的基于整个库的不完全恢复，提高了数据库的可用性和恢复的效率。**

本文出自 “[天涯客的blog][blog]” 博客，请务必保留此出处[http://tiany.blog.51cto.com/513694/1609109][http_tiany.blog.51cto.com_513694_1609109]

[wKiom1TIXrnRCEMWAAGW05m6DzE496.jpg]: /images/20220805/b68b0a4ce2f94d06afe87e061897a930.png
[blog]: http://tiany.blog.51cto.com
[http_tiany.blog.51cto.com_513694_1609109]: http://tiany.blog.51cto.com/513694/1609109