SQL> archive log list;

2.创建归档目录

mkidr /oradata/arch

3.开启归档

SQL> alter databasemount;

SQL>alter databasearchivelog;

SQL>alter system switch logfile;

4.修改主库属性

SQL> alter databaseforce logging;

SQL>select FORCE_LOGGING from v$database;

5.在主库创建密码文件、参数文件、以及standby控制文件

SQL> create pfile='/home/oracle/pfile.ora' from spfile;

6.密码文件可以直接用现有的

ls $ORACLE_HOME/dbs/orapw*

7.创建standby控制文件

SQL> alter database create standby controlfile as '/home/oracle/controlstd.ctl';

8.主库TNS配置

TEST=(DESCRIPTION=(ADDRESS= (PROTOCOL = TCP)(HOST = 10.63.230.10)(PORT = 1521))

(CONNECT_DATA=(SERVER=DEDICATED)

(SERVICE_NAME=TEST)

)

)

TESTBAK=(DESCRIPTION=(ADDRESS= (PROTOCOL = TCP)(HOST = 10.63.230.13)(PORT = 1521))

(CONNECT_DATA=(SERVER=DEDICATED)

(SERVICE_NAME=TESTBAK)

)

)

9.主库备份数据库

RMAN>run {

crosscheck backup ;

crosscheck archivelog all;

crosscheck backupset ;

delete noprompt obsolete device type disk;

delete noprompt expired backup device type disk;

backup as compressed backupset full database format'/oradata/rman/full_%d_%t_%s_%p' plus archivelog format '/oradata/rman/arch_%d_%t_%s_%p.bkp';

}

10.备份完毕后将rman备份文件，standby控制文件，密码文件，参数文件一起传到备库

参数文件和standby控制文件放在/home/oracle/目录下

密码文件放在 $ORACLE_HOME/dbs/目录下

注：如果目录属主不是 oracle.oinstall或者oracle.dba 需要修改目录权限，文件传完后，检查下文件的用户和组。

--备库

1.创建数据库目录,根据自己的参数文件中的参数目录而定

mkdir -p /u01/app/oracle/admin/test/adump

mkdir -p /oradata/rmanmkdir -p /oradata/arch

mkdir -p /oradata/test

2.使用参数文件开启数据库

SQL> startup nomount pfile='/home/oracle/pfile.ora';

3.导入standby控制文件

RMAN> restore controlfile from '/home/oracle/controlstd.ctl';

4.恢复数据文件

SQL> alter database mount;

RMAN> catalog start with '/oradata/rman/';

RMAN> restore database;

5.恢复归档文件

RMAN> recover database;

6.备库TNS配置

TEST=(DESCRIPTION=(ADDRESS= (PROTOCOL = TCP)(HOST = 10.63.230.10)(PORT = 1521))

(CONNECT_DATA=(SERVER=DEDICATED)

(SERVICE_NAME=TEST)

)

)

TESTBAK=(DESCRIPTION=(ADDRESS= (PROTOCOL = TCP)(HOST = 10.63.230.13)(PORT = 1521))

(CONNECT_DATA=(SERVER=DEDICATED)

(SERVICE_NAME=TESTBAK)

)

)

7.创建standby redo log

在使用最大保护模式或是最大可用模式的时候需要在standby数据库创建standby redo log，对于最大性能模式可以不用创建standby redo log

SQL> alter database addstandby logfilegroup 4 ('/oradata/test/std_redo04a.log') size 100m,group 5 ('/oradata/test/std_redo05a.log') size 100m,group 6 ('/oradata/test/std_redo06a.log') size 100m,group 7 ('/oradata/test/std_redo07a.log') size 100m;

SQL> alter database clear logfile group 1;

SQL> alter database clear logfile group 2;

SQL> alter database clear logfile group 3;

添加standby日志文件的规则：

standby日志最少应该比redo log 多一组。推荐的备重做日志数依赖于主数据库上的线程数。

(每线程日志文件最大数目 + 1 ) * 线程数

否则standby日志在应用时报如下信息：

RFS[1]: No standby redo logfiles created

RFS[1]: Archived Log: '/oradata/arch/1_30_633287861.dbf'

在主库添加完standby logfile后，当主库切换为standby角色后会自动使用standby redo logfile，具体应用信息如下：

RFS[1]: Successfully opened standby log 4: '/oradata/test/std_redo04a.log'

修改主备库参数

主库primary

SQL> alter system set db_unique_name='test' scope=spfile;

SQL> alter system set fal_server='testbak' sid='*'scope=both;

SQL> alter system set fal_client='test' sid='*' scope=both;

SQL> alter system set standby_file_management=auto sid='*' scope=both;

SQL> alter system set log_archive_config='dg_config=(test,testbak)' sid='*' scope=both;

SQL> alter system set log_archive_dest_2='service=testbak lgwr async valid_for=(online_logfiles,primary_role) db_unique_name=testbak' sid='*' scope=both;

SQL>alter system set log_archive_dest_state_2=enable;

备库standby

SQL> alter system set db_unique_name='testbak' scope=spfile;

SQL> alter system set fal_server='test' sid='*'scope=both;

SQL> alter system set fal_client='testbak' sid='*' scope=both;

SQL> alter system set standby_file_management=auto sid='*' scope=both;

SQL> alter system set log_archive_config='dg_config=(test,testbak)' sid='*' scope=both;

SQL> alter system set log_archive_dest_2='service=test lgwr async valid_for=(online_logfiles,primary_role) db_unique_name=test' sid='*' scope=both;

SQL> alter system set log_archive_dest_state_2=enable;

开启DG(10g不open)

SQL> alter database open;

SQL> alter database recover managed standby database disconnect from session using currentlogfile;

SQL> select process,client_process,sequence#,status from v$managed_standby;

停用DG

SQL> alter database recover managed standby database cancel;

如果主库传不过来归档，可以通过在主库侧手工修改参数如下：

SQL> ALTER SYSTEM SET LOG_ARCHIVE_DEST_STATE_2='DEFER' SCOPE=MEMORY;

SQL> ALTER SYSTEM SET LOG_ARCHIVE_DEST_STATE_2='ENABLE' SCOPE=MEMORY;

至此dg搭建完毕

查询dg延迟

SQL> col value fora20

SQL> col time_computedfora30

SQL> col namefora25

SQL>col unitfora30

SQL>col datum_timefora30

SQL>set line 400

SQL>select * from v$dataguard_stats;