使用存储脚本还原恢复数据库

如果要使用存储脚本，首先需要启用恢复目录。

1. 创建存储脚本

点击(此处)折叠或打开

1. #RMAN连接到恢复目录
   
2. rman target / catalog rcowner/oracle@rcata
   
3. 
   
4. #创建备份用脚本
5. create script bak_db_arch
6. comment 'backup database plus archivelog, then clean archivelog.'
   {
   allocate channel c1 device type disk;
   allocate channel c2 device type disk;
   backup database plus archivelog;
   delete noprompt archivelog all;
   release channel c1;
   release channel c2;
   }
   
7. #创建还原脚本
8. create script restore_db
9. comment 'restore database, then db can be recovered.'
   {
   restore spfile;
   shutdown immediate;
   startup nomount;
   restore controlfile;
   alter database mount;
   allocate channel c1 device type disk;
   allocate channel c2 device type disk;
   restore database;
   release channel c1;
   release channel c2;
   }
   
10. #创建恢复脚本，需要用到替换变量
11. create script recover_db
    {
    set until sequence &1 thread &2;
    recover database;
    alter database open resetlogs;
    }
    
12. #注意：在有替换变量的这一行，RMAN会立即要求输入初始值，因此不能将整段语句直接粘贴上去！

以下是一些与存储脚本有关的命令

点击(此处)折叠或打开

1. # 查看恢复目录里有哪些脚本，同时会显示comment
   
2. list script names;
   
3. 
   
4. # 查看脚本内容
   
5. print script <脚本名>;
   
6. 
   
7. # 编辑脚本
   
8. replace script ...（...与创建命令相同）
   
9. 
   
10. # 删除脚本
    
11. delete script <脚本名>;

接下来，先执行备份

点击(此处)折叠或打开

1. run { execute script bak_db_arch; }

然后故意破坏参数文件

点击(此处)折叠或打开

1. mv spfileprod.ora spfileprod.ora.old

在RMAN中启动到nomount

点击(此处)折叠或打开

1. RMAN> startup nomount

然后执行restore脚本

点击(此处)折叠或打开

1. RMAN> run { execute script restore_db; }

完成后，查看日志序列号和线程号，这个将在恢复时用到。

点击(此处)折叠或打开

1. RMAN> list backup by file;
   
2. 
   
3. List of Archived Log Backups
   
4. ============================
   
5. 
   
6. Thrd Seq Low SCN Low Time BS Key S #Pieces #Copies Compressed Tag
   
7. ---- ------- ---------- ------------------- ------- - ------- ------- ---------- ---
   
8. 1 2 4471363 2016-04-02 09:00:53 1104 A 1 1 NO TAG20160407T084606
   
9. 1 3 4532043 2016-04-04 01:00:10 1104 A 1 1 NO TAG20160407T084606
   
10. 1 4 4607063 2016-04-06 00:30:18 1105 A 1 1 NO TAG20160407T084606
    
11. 1 1 4625382 2016-04-06 12:55:14 1106 A 1 1 NO TAG20160407T084606
    
12. 1 1 4626459 2016-04-06 13:20:54 1107 A 1 1 NO TAG20160407T084606
    
13. 1 2 4656086 2016-04-07 08:46:00 1142 A 1 1 NO TAG20160407T085111

执行恢复脚本

点击(此处)折叠或打开

1. RMAN> print script recover_db;
   
2. 
   
3. printing stored script: recover_db
   
4. {
   
5. set until sequence &1 thread &2;
   
6. recover database;
   
7. alter database open resetlogs;
   
8. }
   
9. 
   
10. RMAN> run { execute script recover_db; }
    
11. 
    
12. executing script: recover_db
    
13. 
    
14. Enter value for 1: 3
    
15. 
    
16. Enter value for 2: 1
    
17. 
    
18. executing command: SET until clause
    
19. 
    
20. Starting recover at 2016-04-07 09:17:40
    
21. allocated channel: ORA_DISK_1
    
22. channel ORA_DISK_1: SID=63 device type=DISK
    
23. 
    
24. starting media recovery
    
25. 
    
26. archived log for thread 1 with sequence 2 is already on disk as file /u01/app/oracle/oradata/prod/disk1/redo02a.log
    
27. archived log file name=/u01/app/oracle/oradata/prod/disk1/redo02a.log thread=1 sequence=2
    
28. media recovery complete, elapsed time: 00:00:00
    
29. Finished recover at 2016-04-07 09:17:46
    
30. 
    
31. database opened
    
32. new incarnation of database registered in recovery catalog
    
33. starting full resync of recovery catalog
    
34. full resync complete

完成！

来自 “ ITPUB博客 ” ，链接：http://blog.itpub.net/22621861/viewspace-2076680/，如需转载，请注明出处，否则将追究法律责任。

转载于:http://blog.itpub.net/22621861/viewspace-2076680/