课程第31/32次（2016-12-28星期三，2016-12-30星期五）

botang

上完1Z0-053第2章（复习可恢复性）
（052共19章，053共21章，063多租户共9章，49-18）

如果CONFIGURE RETENTION POLICY TO RECOVERY WINDOW OF 1 DAYS; 即24H
那么1T备份做在2016-12-28   21:28:00
2T备份做在2016-12-29 00:14:00
如果现在是2016-12-30 00:14:00，那么1T就obsolete

结论1: obsolete算整整24H

如果在2016-12-30 23:59:59，执行report need backup；没有输出
如果在2016-12-31 00:00:00，执行report need backup；说所有数据文件都需要备份

结论2: report need  backup算24H+当天剩余时间

时间是2016-12-31 00:10:00
如果users表空间只读，并做3T备份
并且CONFIGURE BACKUP OPTIMIZATION ON;

时间到了2016-12-31 23:53:00
做4T备份，发现 users表空间被跳过不备份

最后时间到了2017-01-01 00:00:00
做5T备份，发现 users表空间有开始备份

结论3: 备份优化有效期算24H-当天剩余时间


恢复目录操作：

1. [oracle@station90 admin]$ rman target  /   catalog u90/oracle_4U@utforcl
   
2. 
   
3. Recovery Manager: Release 11.2.0.1.0 - Production on Mon Jan 2 23:36:43 2017
   
4. 
   
5. Copyright (c) 1982, 2009, Oracle and/or its affiliates.  All rights reserved.
   
6. 
   
7. connected to target database: ORCL (DBID=1343950367)
   
8. connected to recovery catalog database
   
9. 
   
10. RMAN> register database ;
    
11. 
    
12. database registered in recovery catalog
    
13. starting full resync of recovery catalog
    
14. full resync complete
    
15. 
    
16. RMAN> unregister database ;
    
17. 
    
18. database name is "ORCL" and DBID is 1343950367
    
19. 
    
20. Do you really want to unregister the database (enter YES or NO)? YES
    
21. database unregistered from the recovery catalog
    
22. 
    
23. RMAN> drop catalog;
    
24. 
    
25. recovery catalog owner is U90
    
26. enter DROP CATALOG command again to confirm catalog removal
    
27. 
    
28. RMAN> drop catalog;
    
29. 
    
30. ORACLE error from recovery catalog database: ORA-00054: resource busy and acquire with NOWAIT specified or timeout expired
    
31. 
    
32. RMAN-00571: ===========================================================
    
33. RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
    
34. RMAN-00571: ===========================================================
    
35. RMAN-06434: some errors occurred while removing recovery catalog
    
36. 
    
37. RMAN> drop catalog;
    
38. 
    
39. recovery catalog owner is U90
    
40. enter DROP CATALOG command again to confirm catalog removal
    
41. 
    
42. RMAN> drop catalog;
    
43. 
    
44. recovery catalog dropped
    
45. 
    
46. RMAN> create catalog;
    

复制代码

做过的实验：


恢复0级镜像拷贝的脚本（图形界面建议的脚本）：

1. run {
   
2. allocate channel c1 device type disk;
   
3. recover copy of database with tag 'ORA_OEM_LEVEL_0';
   
4. backup incremental level 1 cumulative  copies=1 for recover of copy with tag 'ORA_OEM_LEVEL_0' database;
   
5. release channel c1;
   
6. allocate channel c1 type 'SBT_TAPE' format '%U' parms 'ENV=(OB_MEDIA_FAMILY=station90)';
   
7. allocate channel c2 type 'SBT_TAPE' format '%U' parms 'ENV=(OB_MEDIA_FAMILY=station90)';
   
8. backup archivelog all not backed up;
   
9. }
   
10. allocate channel for maintenance device type 'SBT_TAPE' parms 'ENV=(OB_MEDIA_FAMILY=station90)';
    
11. delete noprompt obsolete redundancy 1 device type 'SBT_TAPE';

复制代码