首页 > 代码库 > DataGuard开启failover

DataGuard开启failover

2024-09-09 09:42:49   217人阅读


1、修改保护模式

  1. DGMGRL> edit configuration set protection mode as maxAvailability;
  2. Succeeded.

2、修改日志同步方式

  1. DGMGRL> show database verbose xiaohe ‘LogXptMode‘
  2.   LogXptMode = ‘ASYNC‘
  3. DGMGRL> show database verbose xiaohedg ‘LogXptMode‘
  4.   LogXptMode = ‘ASYNC‘
  5. DGMGRL> edit database xiaohe set property LogXptMode = ‘SYNC‘;
  6. Property "logxptmode" updated
  7. DGMGRL> edit database xiaohedg set property LogXptMode = ‘SYNC‘;
  8. Property "logxptmode" updated
  9. DGMGRL> 
  10. DGMGRL> show database verbose xiaohe ‘LogXptMode‘
  11.   LogXptMode = ‘SYNC‘
  12. DGMGRL> show database verbose xiaohedg ‘LogXptMode‘
  13.   LogXptMode = ‘SYNC‘

3、在dgmgrl中配置主库与standby数据库的FastStartFailoverTarget属性

  1. DGMGRL> show database verbose xiaohe ‘FastStartFailoverTarget‘;
  2.   FastStartFailoverTarget = ‘xiaohedg‘
  3. DGMGRL> show database verbose xiaohedg ‘FastStartFailoverTarget‘;
  4.   FastStartFailoverTarget = ‘xiaohe‘

这里已经不用配置了， 如果需要配置， 使用如下的命令：
edit database xiaohedg set property FastStartFailoverTarget=‘xiaohe‘;

4、启动observer

DGMGRL> start observer
Observer started


5、在另外一个终端查看failover状态

可以看到Observer: xiaohe_dg1， 这里监控的就是主库。

  1. DGMGRL> show fast_start failover
  3. Fast-Start Failover: DISABLED
  5.   Threshold:        30 seconds
  6.   Target:           (none)
  7.   Observer:         xiaohe_dg1
  8.   Lag Limit:        30 seconds
  9.   Shutdown Primary: TRUE
  10.   Auto-reinstate:   TRUE
  12. Configurable Failover Conditions
  13.   Health Conditions:
  14.     Corrupted Controlfile          YES
  15.     Corrupted Dictionary           YES
  16.     Inaccessible Logfile            NO
  17.     Stuck Archiver                  NO
  18.     Datafile Offline               YES
  20.   Oracle Error Conditions:
  21.     (none)


6、激活failover功能

  1. DGMGRL> enable fast_start failover;
  2. Enabled.

7、模拟故障

（1）、将主库给shutdown abort
  1. SQL> shutdown abort;
  2. ORACLE instance shut down.

（2）、等一两分钟， 查看observer的日志
  1. DGMGRL> start observer
  2. Observer started
  5. 23:56:08.19  Saturday, March 18, 2017
  6. Initiating Fast-Start Failover to database "xiaohe"...
  7. Performing failover NOW, please wait...
  8. Failover succeeded, new primary is "xiaohe"
  9. 23:56:11.64  Saturday, March 18, 2017


（3）、查看备库是否已经切换为主库
  1. SQL> select open_mode from v$database;
  3. OPEN_MODE
  4. --------------------
  5. READ WRITE

（4）、将故障的库重新启动
启动数据库报错：
  1. SQL> startup
  2. ORACLE instance started.
  4. Total System Global Area  839282688 bytes
  5. Fixed Size		    2233000 bytes
  6. Variable Size		  494931288 bytes
  7. Database Buffers	  339738624 bytes
  8. Redo Buffers		    2379776 bytes
  9. Database mounted.
  10. ORA-16649: possible failover to another database prevents this database from
  11. being opened



查看日志，下面的红色部分是新产生的日志：

  1. DGMGRL> start observer
  2. Observer started
  5. 23:56:08.19  Saturday, March 18, 2017
  6. Initiating Fast-Start Failover to database "xiaohe"...
  7. Performing failover NOW, please wait...
  8. Failover succeeded, new primary is "xiaohe"
  9. 23:56:11.64  Saturday, March 18, 2017
  11. 00:05:46.24  Sunday, March 19, 2017
  12. Initiating reinstatement for database "xiaohedg"...
  13. Reinstating database "xiaohedg", please wait...
  14. Operation requires shutdown of instance "xiaohedg" on database "xiaohedg"
  15. Shutting down instance "xiaohedg"...
  16. ORA-01109: database not open
  18. Database dismounted.
  19. ORACLE instance shut down.
  20. Operation requires startup of instance "xiaohedg" on database "xiaohedg"
  21. Starting instance "xiaohedg"...
  22. ORACLE instance started.
  23. Database mounted.
  24. Continuing to reinstate database "xiaohedg" ...
  25. Reinstatement of database "xiaohedg" succeeded
  26. 00:06:42.78  Sunday, March 19, 2017

等了几分钟后， 数据库自动恢复正常了。

经测试， 但故障的库只需要启动到mount状态， failover就能自动将库打开。


















DataGuard开启failover


联系
我们