如楼上所说，有三种模式，但一般情况下备库都是默认的最大性能模式，重启备库不会影响主库，但要记得重启后备库要应用日志进程 MRP0.

 select DATABASE_ROLE,PROTECTION_MODE,PROTECTION_LEVEL from v$database;

DATABASE_ROLE    PROTECTION_MODE      PROTECTION_LEVEL
---------------- -------------------- --------------------
PHYSICAL STANDBY MAXIMUM PERFORMANCE  MAXIMUM PERFORMANCE

一般流程如下：

1、先停备库应用
查看备库保护模式
select DATABASE_ROLE,PROTECTION_MODE,PROTECTION_LEVEL from v$database;
====================================================================

2、RAC ADG备库  停备库
先听监听
lsnrctl stop
sqlplus / as sysdba
查看同步情况
set linesize 150; 
set pagesize 9999; 
column name format a13; 
column value format a20; 
column unit format a30; 
column TIME_COMPUTED format a30; 
select name,value,unit,time_computed from v$dataguard_stats where name in ('transport lag','apply lag');

select open_mode from v$database;
select process,status,sequence# from v$managed_standby;

关库
shutdown immediate;
exit;

tail -500f alert_jiekedb.log

3.关机 
shutdown -h now
4.维护完毕，开机

5.启动 ADG 备库

先查看是否开启自启动
su - oracle
ps -ef | grep smon 
lsnrctl status
sqlplus / as sysdba

如果已启动，检查启动到哪个阶段
col HOST_NAME for a30 
select INSTANCE_NAME,HOST_NAME,VERSION,STARTUP_TIME,STATUS from gv$instance; 

如果没有启动,则先启动到 mount 

sqlplus / as sysdba
startup mount 
select open_mode from v$database;

检查 mrp0 进程是否开启
select process,status,sequence# from v$managed_standby;
检查 ADG 同步情况
set linesize 150; 
set pagesize 9999; 
column name format a13; 
column value format a20; 
column unit format a30; 
column TIME_COMPUTED format a30; 
select name,value,unit,time_computed from v$dataguard_stats where name in ('transport lag','apply lag');

如果没有开启则先开启 mrp0 进程，使用如下语句应用日志进程 mrp0
alter database recover managed standby database using current logfile disconnect;

再次查看等如下 value 值为0
set linesize 150; 
set pagesize 9999; 
column name format a13; 
column value format a20; 
column unit format a30; 
column TIME_COMPUTED format a30; 
select name,value,unit,time_computed from v$dataguard_stats where name in ('transport lag','apply lag');

取消日志应用
alter database recover managed standby database cancel;
然后 open 数据库实例

alter database open;

select open_mode from v$database;

然后需要继续应用日志进程 mrp0 
alter database recover managed standby database using current logfile disconnect;
检查 mrp0 进程
select process,status,sequence#,THREAD# from v$managed_standby;

--查看日志应用时间，注意 apply lag 即可。
set linesize 150; 
set pagesize 9999; 
column name format a13; 
column value format a20; 
column unit format a30; 
column TIME_COMPUTED format a30; 
select name,value,unit,time_computed from v$dataguard_stats where name in ('transport lag','apply lag');

exit;

也可查看备库 alter 日志。

验证：
主库 切换日志检查备库日志是否会立即同步
alter system switch logfile;

alter system archive log current;
备库
查看Alert日志更新情况
tail -500f alert_jiekedb.log