Oracle恢复实验（一）

环境：Oracle10g、Red Hat 4，Oracle运行在归档模式。

模拟场景：user01.dbf数据文件损坏，关闭数据库条件下恢复。

具体步骤：

1、先做个全库冷备份，备份控制文件、数据文件、参数文件

SQL> select name from v$controlfile

2 union

3 select name from v$datafile;

NAME

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

/home/oracle/oracle/product/oradata/orcl/control01.ctl

/home/oracle/oracle/product/oradata/orcl/control02.ctl

/home/oracle/oracle/product/oradata/orcl/control03.ctl

/home/oracle/oracle/product/oradata/orcl/sysaux01.dbf

/home/oracle/oracle/product/oradata/orcl/system01.dbf

/home/oracle/oracle/product/oradata/orcl/undotbs01.dbf

/home/oracle/oracle/product/oradata/orcl/users01.dbf

7 rows selected.

　也就是关库，备份上面几个文件，并且备份参数文件，口令文件可备可不备。

2、用scott用户登录，建个表，插入些数据。提交，检查一下数据行数。

SQL> conn scott/oracle

Connected.

SQL> create table recover_test tablespace users as select * from emp;

Table created.

SQL> insert into recover_test (select * from emp);

14 rows created.

SQL> insert into recover_test select * from recover_test;

56 rows created.

SQL> /

112 rows created.

SQL> /

224 rows created.

SQL> commit;

Commit complete.

SQL> select count (*) from recover_test;

COUNT(*)

----------

448

3、多做几次日志切换，这样有且于了解Oracle恢复的原理。

SQL> conn /as sysdba

Connected.

SQL> alter system switch logfile;

System altered.

SQL> /

System altered.

SQL> /

System altered.

SQL> /

System altered.

SQL> /

System altered.

SQL> /

System altered.

4、主机断电，删除user01.dbf，模拟数据文件损坏

SQL> shutdown abort

ORACLE instance shut down.

SQL> !rm -f /home/oracle/oracle/product/oradata/orcl/user01.dbf

5、尝试打开数据库

SQL> startup

ORACLE instance started.

Total System Global Area 130023424 bytes

Fixed Size 1218100 bytes

Variable Size 62917068 bytes

Database Buffers 62914560 bytes

Redo Buffers 2973696 bytes

Database mounted.

ORA-01157: cannot identify/lock data file 4 - see DBWR trace file

ORA-01110: data file 4: '/home/oracle/oracle/product/oradata/orcl/users01.dbf'

此时提示数据文件找不到。警告日志中可以看到更加详细的信息。

6、恢复数据文件，尝试打开数据库

SQL> !cp /disk2/bak/users01.dbf /home/oracle/oracle/product/oradata/orcl/

SQL> alter database open;

alter database open

*

ERROR at line 1:

ORA-01113: file 4 needs media recovery

ORA-01110: data file 4: '/home/oracle/oracle/product/oradata/orcl/users01.dbf'

此时由于数据库不一致，提示需要介质恢复。检查一下scn情况。

select 'scnOfDatafile',name,checkpoint_change# from v$datafile

union

select 'scnOfDatafileHeader',name,checkpoint_change# from v$datafile_header;

'SCNOFDATAFILE' NAME CHECKPOINT_CHANGE#

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

scnOfDatafile /home/oracle/oracle/product/oradata/orcl/sysaux01.dbf 743872

scnOfDatafile /home/oracle/oracle/product/oradata/orcl/system01.dbf 743872

scnOfDatafile /home/oracle/oracle/product/oradata/orcl/undotbs01.dbf 743872

scnOfDatafile /home/oracle/oracle/product/oradata/orcl/users01.dbf 743872

scnOfDatafileHeader /home/oracle/oracle/product/oradata/orcl/sysaux01.dbf 743872

scnOfDatafileHeader /home/oracle/oracle/product/oradata/orcl/system01.dbf 743872

scnOfDatafileHeader /home/oracle/oracle/product/oradata/orcl/undotbs01.dbf 743872

scnOfDatafileHeader /home/oracle/oracle/product/oradata/orcl/users01.dbf 721971

9 rows selected.

可以明显的看到数据文件users01.dbf的scn比较小，这是备份时的scn。

7、介质恢复

SQL> recover database;

ORA-00279: change 721971 generated at 09/25/2011 13:27:36 needed for thread 1

ORA-00289: suggestion : /home/oracle/oracle/product/archive/1_7_760842868.dbf

ORA-00280: change 721971 for thread 1 is in sequence #7

Specify log: {<RET>=suggested | filename | AUTO | CANCEL}

auto

ORA-00279: change 723634 generated at 09/25/2011 14:05:56 needed for thread 1

ORA-00289: suggestion : /home/oracle/oracle/product/archive/1_8_760842868.dbf

ORA-00280: change 723634 for thread 1 is in sequence #8

ORA-00278: log file '/home/oracle/oracle/product/archive/1_7_760842868.dbf' no longer needed for this recovery

ORA-00279: change 723636 generated at 09/25/2011 14:05:57 needed for thread 1

ORA-00289: suggestion : /home/oracle/oracle/product/archive/1_9_760842868.dbf

ORA-00280: change 723636 for thread 1 is in sequence #9

ORA-00278: log file '/home/oracle/oracle/product/archive/1_8_760842868.dbf' no longer needed for this recovery

ORA-00279: change 743752 generated at 09/25/2011 14:10:32 needed for thread 1

ORA-00289: suggestion : /home/oracle/oracle/product/archive/1_10_760842868.dbf

ORA-00280: change 743752 for thread 1 is in sequence #10

ORA-00278: log file '/home/oracle/oracle/product/archive/1_9_760842868.dbf' no longer needed for this recovery

ORA-00279: change 743864 generated at 09/25/2011 14:10:58 needed for thread 1

ORA-00289: suggestion : /home/oracle/oracle/product/archive/1_11_760842868.dbf

ORA-00280: change 743864 for thread 1 is in sequence #11

ORA-00278: log file '/home/oracle/oracle/product/archive/1_10_760842868.dbf' no longer needed for this recovery

ORA-00279: change 743866 generated at 09/25/2011 14:10:59 needed for thread 1

ORA-00289: suggestion : /home/oracle/oracle/product/archive/1_12_760842868.dbf

ORA-00280: change 743866 for thread 1 is in sequence #12

ORA-00278: log file '/home/oracle/oracle/product/archive/1_11_760842868.dbf' no longer needed for this recovery

ORA-00279: change 743868 generated at 09/25/2011 14:11:05 needed for thread 1

ORA-00289: suggestion : /home/oracle/oracle/product/archive/1_13_760842868.dbf

ORA-00280: change 743868 for thread 1 is in sequence #13

ORA-00278: log file '/home/oracle/oracle/product/archive/1_12_760842868.dbf' no longer needed for this recovery

使用auto选项进行恢复，可以看到使用了8-13号日志，到操作系统中看一下：

[oracle@localhost archive]$ ls -lt

total 4112

-rw-r----- 1 oracle oinstall 4608 Sep 25 14:11 1_15_760842868.dbf

-rw-r----- 1 oracle oinstall 1024 Sep 25 14:11 1_14_760842868.dbf

-rw-r----- 1 oracle oinstall 1024 Sep 25 14:11 1_13_760842868.dbf

-rw-r----- 1 oracle oinstall 2048 Sep 25 14:11 1_12_760842868.dbf

-rw-r----- 1 oracle oinstall 1024 Sep 25 14:10 1_11_760842868.dbf

-rw-r----- 1 oracle oinstall 53760 Sep 25 14:10 1_10_760842868.dbf

-rw-r----- 1 oracle oinstall 1536 Sep 25 14:10 1_8_760842868.dbf

-rw-r----- 1 oracle oinstall 30720 Sep 25 14:10 1_9_760842868.dbf

-rw-r----- 1 oracle oinstall 4043776 Sep 25 14:10 1_7_760842868.dbf

还有14、15两个归档日志，可这两个并没有使用。这是为什么？

查看一下警告日志

[oracle@localhost bdump]$ tail -f alert_orcl.log

Mem# 0 errs 0: /home/oracle/oracle/product/oradata/orcl/redo01.log

Sun Sep 25 14:33:19 2011

Recovery of Online Redo Log: Thread 1 Group 2 Seq 15 Reading mem 0

Mem# 0 errs 0: /home/oracle/oracle/product/oradata/orcl/redo02.log

Sun Sep 25 14:33:19 2011

Recovery of Online Redo Log: Thread 1 Group 3 Seq 16 Reading mem 0

Mem# 0 errs 0: /home/oracle/oracle/product/oradata/orcl/redo03.log

Sun Sep 25 14:33:19 2011

Media Recovery Complete (orcl)

Completed: ALTER DATABASE RECOVER CONTINUE DEFAULT

根据以上信息，可以看到：Oracle使用的redo日志。

结论：Oracle介质恢复，使用归档日志+Redo日志，优先使用Redo日志。应用完Redo日志，数据库就可以打开，如果需要Undo日志，继续Undo日志做回滚。可以在第2步发起一个未提交事务，abort关库后，可以看到效果。

8、验证一下SCN

select 'scnOfDatafile',name,checkpoint_change# from v$datafile

union

select 'scnOfDatafileHeader',name,checkpoint_change# from v$datafile_header

union

select 'scnOfDatabase','database',checkpoint_change# from v$database;

'SCNOFDATAFILE' NAME CHECKPOINT_CHANGE#

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

scnOfDatafile /home/oracle/oracle/product/oradata/orcl/sysaux01.dbf 743906

scnOfDatafile /home/oracle/oracle/product/oradata/orcl/system01.dbf 743906

scnOfDatafile /home/oracle/oracle/product/oradata/orcl/undotbs01.dbf 743906

scnOfDatafile /home/oracle/oracle/product/oradata/orcl/users01.dbf 743906

scnOfDatafileHeader /home/oracle/oracle/product/oradata/orcl/sysaux01.dbf 743906

scnOfDatafileHeader /home/oracle/oracle/product/oradata/orcl/system01.dbf 743906

scnOfDatafileHeader /home/oracle/oracle/product/oradata/orcl/undotbs01.dbf 743906

scnOfDatafileHeader /home/oracle/oracle/product/oradata/orcl/users01.dbf 743906

8 rows selected.

发现SCN号已经一致。此时数据库已经处于同步状态。

9、检查数据是否恢复

SQL> select count(*) from scott.recover_test;

COUNT(*)

----------

448