探索Oracle之RMAN_07单个数据文件丢失恢复

1、 数据文件的丢失恢复

1.1 在wwl表空间上创建5张表，并添加数据。

SQL> create table wwl01 (id number(3),namevarchar2(10));

Table created.

SQL> insert into wwl01 values(1,'wwl');

1 row created.

SQL> insert into wwl01 values(2,'wm');

1 row created.

SQL> insert into wwl01 values(3,'zq');

1 row created.

SQL> insert into wwl01 values(4,'wbq');

1 row created.

SQL> insert into wwl01 values(5,'wq');

1 row created.

SQL> create table wwl02 as select * from wwl01;

Table created.

SQL> create table wwl03 as select * from wwl01;

Table created.

SQL> create table wwl04 as select * from wwl01;

Table created.

SQL> create table wwl05 as select * from wwl01;

Table created.

查看表中的数据：

SQL> select * from tab;

TNAME TABTYPE CLUSTERID

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

WWL01 TABLE

WWL02 TABLE

WWL03 TABLE

WWL04 TABLE

WWL05 TABLE

1.2 执行全库备份

[oracle@wwldb ~]$ rman target /

Recovery Manager: Release 10.2.0.1.0 - Production on Fri Jun 2200:59:59 2012

Copyright (c) 1982, 2005, Oracle. All rights reserved.

connected to target database: WWL (DBID=5520179)

RMAN> backup database;

1.3 模拟数据丢失，手动删除数据文件wwl001.dbf

[oracle@wwldb WWL]$ rm -rfwwl001.dbf

1.4 再次启动数据库，无法启动并报错不能锁定数据文件5，查看dbwr的跟踪文件。

SQL>startup force;

ORACLEinstance started.

TotalSystem Global Area 285212672 bytes

FixedSize 1218968 bytes

VariableSize 92276328 bytes

DatabaseBuffers 184549376 bytes

RedoBuffers 7168000 bytes

Databasemounted.

ORA-01157:cannot identify/lock data file 5 - see DBWR trace file

ORA-01110: data file 5:'/DBData/WWL/wwl001.dbf'

1.5 检查跟踪文件，报如下错误，非常的清楚的告诉了找不到的文件：

Errors in file /DBSoft/admin/WWL/bdump/wwl_dbw0_29185.trc:

ORA-01157: Message 1157 not found; No messagefile for product=RDBMS, facility=ORA; arguments: [5]

ORA-01110: Message 1110 not found; No messagefile for product=RDBMS, facility=ORA; arguments: [5] [/DBData/WWL/wwl001.dbf]

ORA-27037: Message 27037 not found; No messagefile for product=RDBMS, facility=ORA

Linux Error: 2: No such file or directory

Additional information: 3

ORA-1157 signalled during: ALTER DATABASEOPEN...

由以上信息可以得出数据库故障是由于数据文件wwl001.dbf数据文件异常丢失或者损坏导致数据库的故障，那么数据文件丢失就必然存在数据库数据的丢失，但是万幸的是，在丢失之前我们已经做过备份了，现在我们就来通过之前的备份将wwl001.dbf文件恢复回来，数据文件的恢复分为两种，一直是在不影响数据库其它业务情况下的在线联机恢复，还有一种是停机停业务的恢复，详细见如下：

1.6 恢复方法一，零停机，在线恢复

开始执行恢复操作，分为如下七个步骤：

RMAN>startup force mount;

Oracle instancestarted

databasemounted

Total System GlobalArea 285212672 bytes

Fixed Size 1218968 bytes

VariableSize 92276328 bytes

DatabaseBuffers 184549376 bytes

RedoBuffers 7168000 bytes

SQL> alter database datafile '/DBData/WWL/wwl001.dbf' offline;

Database altered.

3、开启数据库

SQL> alter database open;

Database altered.

RMAN> restore datafile'/DBData/WWL/wwl001.dbf';

Starting restore at 22-JUN-12

using target database control fileinstead of recovery catalog

allocated channel: ORA_DISK_1

channel ORA_DISK_1: sid=141devtype=DISK

channel ORA_DISK_1: startingdatafile backupset restore

channel ORA_DISK_1: specifyingdatafile(s) to restore from backup set

restoring datafile 00005 to /DBData/WWL/wwl001.dbf

channel ORA_DISK_1: reading frombackup piece /DBBak/bak_WWL_06_22_0vne4ph6_1_1

channel ORA_DISK_1: restored backuppiece 1

piecehandle=/DBBak/bak_WWL_06_22_0vne4ph6_1_1 tag=TAG20120622T010021

channel ORA_DISK_1: restorecomplete, elapsed time: 00:00:07

Finished restore at 22-JUN-12

[oracle@wwldb WWL]$ ll wwl*

-rw-r----- 1 oracle oinstall52436992 Jun 22 01:21 wwl001.dbf

-rw-r----- 1 oracle oinstall 5251072 Jun 22 01:20 wwl002.dbf

-rw-r----- 1 oracle oinstall 5251072 Jun 22 01:20 wwl003.dbf

RMAN>recover datafile '/DBData/WWL/wwl001.dbf';

Startingrecover at 22-JUN-12

using targetdatabase control file instead of recovery catalog

allocatedchannel: ORA_DISK_1

channelORA_DISK_1: sid=144 devtype=DISK

starting mediarecovery

media recoverycomplete, elapsed time: 00:00:01

Finishedrecover at 22-JUN-12

RMAN>

6、将数据文件在线

SQL> conn / as sysdba

Connected.

SQL> alter database datafile'/DBData/WWL/wwl001.dbf' online;

Database altered.

7、验证数据是否恢复，可以看到数据全部恢复回来了。

SQL>conn wwl/wwl

Connected.

SQL>select * from tab;

TNAME TABTYPE CLUSTERID

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

WWL01 TABLE

WWL02 TABLE

WWL03 TABLE

WWL04 TABLE

WWL05 TABLE

SQL>select count(*) from tab;

COUNT(*)

----------

5

SQL>

1.7 恢复方法二，离线恢复

开始执行恢复操作，分为五个步骤：

SQL> startupforce mount;

ORACLE instancestarted.

Total SystemGlobal Area 285212672 bytes

Fixed Size 1218968 bytes

VariableSize 96470632 bytes

DatabaseBuffers 180355072 bytes

RedoBuffers 7168000 bytes

Databasemounted.

RMAN> restoredatafile '/DBData/WWL/wwl002.dbf'

2> ;

Starting restoreat 22-JUN-12

using targetdatabase control file instead of recovery catalog

allocatedchannel: ORA_DISK_1

channelORA_DISK_1: sid=156 devtype=DISK

channelORA_DISK_1: starting datafile backupset restore

channelORA_DISK_1: specifying datafile(s) to restore from backup set

restoringdatafile 00006 to /DBData/WWL/wwl002.dbf

channelORA_DISK_1: reading from backup piece /DBBak/bak_WWL_06_22_0vne4ph6_1_1

channelORA_DISK_1: restored backup piece 1

piecehandle=/DBBak/bak_WWL_06_22_0vne4ph6_1_1 tag=TAG20120622T010021

channelORA_DISK_1: restore complete, elapsed time: 00:00:01

Finished restoreat 22-JUN-12

RMAN> recover datafile '/DBData/WWL/wwl002.dbf';

Starting recover at 22-JUN-12

using channel ORA_DISK_1

starting media recovery

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

Finished recover at 22-JUN-12

RMAN>

4、开启数据库

SQL> conn /as sysdba

Connected.

SQL> alterdatabase open;

Databasealtered.

5、验证数据是否恢复

SQL> select *from tab;

TNAME TABTYPE CLUSTERID

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

WWL01 TABLE

WWL02 TABLE

WWL03 TABLE

WWL04 TABLE

WWL05 TABLE

SQL> selectcount(*) from wwl01;

COUNT(*)

----------

5

SQL>