Linux下Oracle 10g DataGuard配置(主从同步及切换)

环境描述:
主库:
 IP: 10.8.3.191
 主机名:primary
ORACLE_SID=afc
ORACLE_BASE=/u01/oracle
ORACLE_HOME=/u01/oracle/10g
备库:
 IP: 10.8.3.192
 主机名:standby
ORACLE_SID=afc
ORACLE_BASE=/u01/oracle
ORACLE_HOME=/u01/oracle/10g
 
准备工作:分别在主库和备库都安装上oracle软件,不装数据库。
安装配置步骤:
 1、运行dbca创建数据库
 2、登陆到数据库上
  [oracle@primary ~]$ sqlplus /nolog
  SQL*Plus: Release 10.2.0.1.0 - Production on 星期四 3月 11 20:31:09 2010
  Copyright (c) 1982, 2005, Oracle.  All rights reserved.
  SQL> conn / as sysdba;
  Connected.
  SQL>
 3、将主库设置为 FORCE LOGGING 模式
  SQL> ALTER DATABASE FORCE LOGGING;
  Database altered.


 4、创建一个密码文件,如果数据库是用dbca创建的则会在$ORACLE_HOME/dbs/下自动创建一个叫orapwdSID的一个密码文件。否则可以用  orapwd命令创建一个。


 5、检查数据库是否处于归档状态
  SQL> archive log list;
  Database log mode              Archive Mode
  Automatic archival             Enabled
  Archive destination            /u01/app/oracle/arch_log
  Oldest online log sequence     1
  Next log sequence to archive   2
  Current log sequence           2
  如果不是处于归档状态则运行下面命令
  SQL> SHUTDOWN IMMEDIATE;
  SQL> STARTUP MOUNT;
  SQL> ALTER DATABASE ARCHIVELOG;
  SQL> ALTER DATABASE OPEN;


 6、创建备库的控制文件(创建后数据库不能做结构性的改变)
  SQL> ALTER DATABASE CREATE STANDBY CONTROLFILE AS '/u01/app/oracle/oradata/dgdemo/stdby.ctl';


 7、创建pfile以便于修改配置参数
  SQL> CREATE PFILE FROM SPFILE;


 8、关掉数据库开始编辑参数
  SQL> shutdown immediate;


 9、$ vi /u01/oracle/10g/dbs/initafc.ora (重点处
  添加下面内容
DB_UNIQUE_NAME=primary
  LOG_ARCHIVE_CONFIG='DG_CONFIG=(primary,standby)'
  LOG_ARCHIVE_DEST_1='location=/u01/oracle/flash_recovery_area
  valid_for=(ALL_LOGFILES,ALL_ROLES)
  DB_UNIQUE_NAME=primary'
  LOG_ARCHIVE_DEST_2=
   'SERVICE=standby LGWR ASYNC
    VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE)
    DB_UNIQUE_NAME=standby'
  LOG_ARCHIVE_DEST_STATE_1=ENABLE
  LOG_ARCHIVE_DEST_STATE_2=ENABLE
  LOG_ARCHIVE_MAX_PROCESSES=30
  FAL_SERVER=standby
  FAL_CLIENT=primary
  DB_FILE_NAME_CONVERT='standby','primary'
  STANDBY_FILE_MANAGEMENT=AUTO


 10、进入sqlplus创建spfile(NOMOUNT状态)
  SQL> CREATE SPFILE from PFILE;


 11、创建tnsnames.ora和listener.ora,可以手动创建也可以用netca创建,无论哪种方法都必须保证能互相畅通。
  $ cd /u01/oracle/10g/network/admin/
  $ vi tnsnames.ora
  输入下面内容
 primary =
    (DESCRIPTION =
   (ADDRESS_LIST =
     (ADDRESS = (PROTOCOL = TCP)(HOST = 10.8.3.191)(PORT = 1521))
   )
   (CONNECT_DATA =
     (SERVER = DEDICATED)
     (SERVICE_NAME = afc)
   )
    )
  standby =
    (DESCRIPTION =
   (ADDRESS_LIST =
     (ADDRESS = (PROTOCOL = TCP)(HOST = 10.8.3.192)(PORT = 1521))
   )
   (CONNECT_DATA =
     (SERVER = DEDICATED)
     (SERVICE_NAME = afc)
   )
    )

 

   $ vi listener.ora

SID_LIST_LISTENER =
  (SID_LIST =
    (SID_DESC =
      (SID_NAME = PLSExtProc)
      (ORACLE_HOME = /u01/oracle/10g)
      (PROGRAM = extproc)
    )
  )

LISTENER =
  (DESCRIPTION_LIST =
    (DESCRIPTION =
      (ADDRESS = (PROTOCOL = TCP)(HOST = primary)(PORT = 1521))
      (ADDRESS = (PROTOCOL = IPC)(KEY = EXTPROC0))
    )
  )
 
  12、复制备份文件到备库
  $ cd /u01/app/oracle/
  $ scp -r flash_recovery_area/ admin/ oradata/ 10.8.3.192:/u01/oracle/
  $ cd /u01/oracle/10g/network/admin/
  $ scp listener.ora  tnsnames.ora 10.8.3.192:/u01/oracle/10g/network/admin/
  $ cd /u01/oracle/10g/dbs/
  $ scp orapwdafc 10.8.3.192:/u01/oracle/10g/dbs/


  13、打开备库做以下操作
  $ cd /u01/oracle/oradata/afc
  $ rm -rf control0*
  $ cp stdby.ctl control01.ctl
  $ cp stdby.ctl control02.ctl
  $ cp stdby.ctl control03.ctl

  $ vi initafc.ora 修改备的参数文件如下(重点处
DB_UNIQUE_NAME=standby
  LOG_ARCHIVE_CONFIG='DG_CONFIG=(primary,standby)'
  LOG_ARCHIVE_DEST_1='location=/u01/oracle/flash_recovery_area
  valid_for=(ALL_LOGFILES,ALL_ROLES)
  DB_UNIQUE_NAME=standby'
  LOG_ARCHIVE_DEST_2=
   'SERVICE=primary LGWR ASYNC
    VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE)
    DB_UNIQUE_NAME=primary'
  LOG_ARCHIVE_DEST_STATE_1=ENABLE
  LOG_ARCHIVE_DEST_STATE_2=ENABLE
  LOG_ARCHIVE_MAX_PROCESSES=30
  FAL_SERVER=primary
  FAL_CLIENT=standby
  DB_FILE_NAME_CONVERT='standby','primary'
  STANDBY_FILE_MANAGEMENT=AUTO
  $ cd /u01/app/oracle/10.2.0/db_1/network/admin/
  $ vi listener.ora
  将primary修改为standby


 13、在备库中进入sqlplus
  $ sqlplus /nolog
  SQL> conn / as sysdba;
  SQL> CREATE SPFILE from PFILE='/u01/app/oracle/oradata/dgdemo/pfile';


 14、测试oracle net是否畅通
  首先启动监听服务(分别在主机和备机上进行)
  $ lsnrctl stop
  $ lsnrctl start
  分别在主机和备机上运行下面命令来测试
  $ tnsping primary
  $ tnsping standby


 15、启动主库(正常启动)
  $ sqlplus /nolog
  SQL> conn / as sysdba;
  SQL> startup


 16、启动备库到mount状态,并打开redo Apply
  $ sqlplus /nolog
  SQL> conn / as sysdba;
  SQL> startup mount
  SQL> ALTER DATABASE RECOVER MANAGED STANDBY DATABASE DISCONNECT FROM SESSION;


 17、在主库上做一次日志切换
  SQL> ALTER SYSTEM SWITCH LOGFILE;

  以下语句可以查看日志应用情况
  SQL> SELECT SEQUENCE#,APPLIED FROM V$ARCHIVED_LOG  ORDER BY SEQUENCE#;
   SEQUENCE# APP
  ---------- ---
     2 YES
     3 YES

 如果出现上面情况则说明配置成功。


 18、在主库建表然后在备库查询进行验证
  SQL> create table haha(id integer,name char(10))
  SQL> insert into haha values(1,'xiao');
  SQL> commit;
  SQL> select * from haha;
    ID NAME
  ---------- ----------
     1 xiao


 19、在主库上做个日志归档
  SQL> ALTER SYSTEM ARCHIVE LOG CURRENT;


 20、在备库上取消掉redo apply
  SQL> ALTER DATABASE RECOVER MANAGED STANDBY DATABASE CANCEL;
  SQL> ALTER DATABASE OPEN;
  SQL> select * from haha;
    ID NAME
  ---------- ----------
     1 xiao
 测试成功


 24、再次切换为备库的redo apply状态
  SQL> ALTER DATABASE RECOVER MANAGED STANDBY DATABASE DISCONNECT FROM SESSION;
 主库备库互相切换


 25、在主库上做如下sql语句
  SQL> ALTER DATABASE COMMIT TO SWITCHOVER TO PHYSICAL STANDBY WITH SESSION SHUTDOWN;
  SQL> shutdown immediate;
  SQL> startup nomount;
  SQL> ALTER DATABASE MOUNT STANDBY DATABASE;
  SQL> RECOVER MANAGED STANDBY DATABASE DISCONNECT


 26、在备库上做switchover
  SQL> ALTER DATABASE COMMIT TO SWITCHOVER TO PRIMARY;
  SQL> shutdown immediate;
  SQL> startup


 27、分别在新的主库和备库上查看状态,验证主库备库是否切换正常。
  SQL> select name,database_role from v$database;
 
 再次创建表在切换后的主库备库上验证


 28、在新的主库上创建列 
 
  SQL> insert into haha values(2,'yang');
  SQL> commit;
  SQL> select * from haha;

        ID NAME
---------- ----------
         1 xiao
         2 yang
  SQL> ALTER SYSTEM ARCHIVE LOG CURRENT;


 29、在新的备库上做下面操作以取消redo apply
  SQL> ALTER DATABASE RECOVER MANAGED STANDBY DATABASE CANCEL;
  SQL> ALTER DATABASE OPEN;
  SQL> select * from haha;

        ID NAME
---------- ----------
         1 xiao
         2 yang


 30、恢复新备库为redo apply状态

     最后将数据库切换出去主备状态,如果相关同步都正常说明配置没问题可以用于生产了;
 
 31、主备切换时候遇到如下错误提示(主数据切换为备数据库后再切换回主系统时出现错误),解决
 如下:
 SQL> alter database commit to switchover to primary;
 alter database commit to switchover to primary
 *
 ERROR at line 1:
 ORA-16139: media recovery required

 mount模式下执行alter database recover managed standby database finish
 SQL> alter database commit to switchover to primary;
 Database altered.

你可能感兴趣的:(dataguard)