常用的在线备库搭建有Rman auxiliary 和Rman备份集恢复两种方式。
（一）
在环境配置OK的问题下，使用一条Rman命令就可以完成备库初始化同步，比较简单。

# 登陆
rman target sys/Oracle123@orcl_pri auxiliary sys/Oracle123@orcl_dg
# 开始搭建辅助数据库
RMAN>duplicate target database for standby nofilenamecheck from active database;

（二）
Rman备份集恢复的步骤相对多一些，但也不是很复杂。
（1）规划数据库参数

DB	db_name	db_unique_name	tnsname
primary	orcl	orcl	orcl_pri
standby	orcl	orcl_s	orcl_dg

DB	db_file_base	log_file_base
primary	/oradata/orcl	/oradata/log
standby	/oradata/orcl_dg	/oradata/log_dg

（2）主库参数配置（不考虑failover，只配置最少的参数）

alter system set log_archive_config = 'dg_config=(orcl,orcl_s)';
alter system set log_archive_dest_2 = 'SERVICE=orcl_dg ASYNC VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=orcl_s' scope=both;
alter system set log_archive_dest_state_2 = ENABLE;
alter system set standby_file_management = auto;
# 生成pfile，作为备库启动的参数文件
create pfile=/home/oracle/pfile.ora' from spfile;

（3）备库直接修改pfile.ora参数，并启动至nomount状态

*.db_unique_name='center_s'
*.log_archive_dest_1='location=/oradata/log_dg'
*.log_archive_dest_2='SERVICE=orcl_pri ASYNC VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=orcl'
*.db_file_name_convert='/oradata/orcl','/oradata/orcl_dg'
*.log_file_name_convert='/oradata/log','/oradata/log_dg'
#  启动数据库
create spfile from pfile='/home/oracle/pfile.ora'
startup nomount

（4）tnsnames.ora 和 listener.ora配置
（5）主库备份，上传至备库

# 全备库
RMAN> run {
 allocate channel t1 type disk;
 backup database FORMAT '/oradata/backup/data_%T_%s'; 
 release channel t1;
}
# 控制文件备份
RMAN> backup current controlfile for standby format '/oradata/backup/control.bak';

（5）备库恢复数据库

# 恢复控制文件
RMAN> RESTORE STANDBY CONTROLFILE FROM '/oradata/backup/control.bak';

# 启动数据库到mount状态
RMAN> sql 'alter database mount'

# 恢复数据库
RMAN> catalog start with '/oradata/backup/';
RMAN> restore database;
RMAN> recover database;
# 恢复有报错，可以忽略
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-03002: failure of recover command at 10/17/2017 09:33:30
RMAN-06054: media recovery requesting unknown archived log for thread 1 with sequence xx  and starting SCN of xxxx

（6）创建备库standby日志文件，用于实时同步数据

alter system set standby_file_management = manual;
ALTER DATABASE ADD standby LOGFILE group  11 '/oradata/log_dg/std11.log' size 500m ;
ALTER DATABASE ADD standby LOGFILE group  12 '/oradata/log_dg/std12.log' size 500m ;
ALTER DATABASE ADD standby LOGFILE group  13 '/oradata/log_dg/std13.log' size 500m ;
ALTER DATABASE ADD standby LOGFILE group  14 '/oradata/log_dg/std14.log' size 500m ;

（7）开始实时日志应用并打开数据库

select open_mode ,flashback_on,current_scn from v$database;
# 应用日志
alter database recover managed standby database disconnect from session;
# 查看应用状态，并在主库切换日志，查看序列变化
select max(sequence#),applied,archived from v$archived_log group by applied,archived;
# 打开数据库
alter database recover managed standby database cancel;
alter database open;
alter database recover managed standby database disconnect from session using current logfile;

Oracle DG是数据库架构中最常用的一种，一定要能熟练搭建DG环境和掌握DG failover /switchover 、recovery的方式。