oracle dg安装

主备库均为归档模式，并且force logging
==开启归档并设为强制日志模式==
select log_mode,force_logging from v$database;
alter database archivelog;
alter database force logging;
===修改主库参数===
alter system set log_archive_config="dg_config=(db01,dg01)" scope=both;
alter system set standby_file_management=auto scope=both;
alter system set log_archive_dest_1="location=/oracle/app/oracle/oradata/arch/ valid_for=(all_logfiles,all_roles) db_unique_name=db01" scope=both;
alter system set log_archive_dest_state_1="enable";
alter system set log_archive_dest_2="service=dg01 lgwr async valid_for=(online_logfile,primary_role) db_unique_name=dg01" scope=both;
alter system set log_archive_dest_state_2="enable";
alter system set fal_server=dg01 scope=both;
alter system set fal_client=db01 scope=both;
alter system set db_file_name_convert="/oracle/app/oracle/oradata/dg01/","/oracle/app/oracle/oradata/db01/" scope=spfile;
alter system set log_file_name_convert="/oracle/app/oracle/oradata/dg01/","/oracle/app/oracle/oradata/db01/" scope=spfile;
===创建参数文件，修改后用于备库===
create pfile='/tmp/tt/initdg01.ora' from spfile;
===tnsnames.ora和监听（两端均为静态监听）===
===复制主库准备好的参数文件和密码文件到备库===
环境准备好后：
===备库利用修改好的参数文件，启动数据库到nomount状态===


===主库操作===
rman target / auxiliary sys/oracle@dg01
duplicate target database for standby from active database;
或者
duplicate target database for standby nonamefilecheck from active database;（无需目录转换）
===建立standy log（redolog+1）===
alter database add standby logfile thread 1 group 11 '/oracle/app/oracle/oradata/dg01/standby_group_11.log' size 52428800;
alter database add standby logfile thread 1 group 12 '/oracle/app/oracle/oradata/dg01/standby_group_12.log' size 52428800;
alter database add standby logfile thread 1 group 13 '/oracle/app/oracle/oradata/dg01/standby_group_13.log' size 52428800;
===dg验证===
验证物理备库是否工作：
检查是否存在报错：
select error,destination from v$archive_dest;
(1)在备库上检查已经存在的归档日志
select sequence# , first_time , next_time from  v$archived_log order  by  sequence#;
(2)在主库上强制进行日志切换，手动归档
alter  system  switch  logfile;
(3)在备库上检查归档日志(是否出现新的归档日志)
select sequence# , first_time , next_time from  v$archived_log order  by  sequence#;
(4)检查归档日志是否被“应用”
select  sequence# , applied  from  v$archived_log order by sequence#;
===当备库正在应用日志，把备库启动到只读模式
  　停掉日志应用
alter database recover managed standby database cancel;
以只读方式打开备库
alter database open;


当数据库处于只读模式，把备库切换到Redo Apply
1. 让所有活动的会话退出备库
2. 日志应用
alter  database  recover  managed  standby  database disconnect  from  session;
或者(启动real-time apply)
alter  database  recover  managed  standby  database using  current  logfile disconnect  from  session;
====主备切换===
1、主库切换到备库
查看主库状态：
select database_role,switchover_status from v$database;
to standby
alter database commit to switchover to physical standby;
备注：
A：switchover_status出现session active/not allow
session active 表示还有活动的session
alter database commit to switchover to physical standby with session shut down;
B:not allow,ora-01153:an incompatible media recovery is active
alter database recover managed standby database finish;
或者
alter database recover managed standby database finish force;
重启主库
shutdown immediate
startup mount 
alter  database  recover  managed  standby  database disconnect  from  session;
select database_role,switchover_status from v$database;
to primary
2、备库切换到主库
select database_role,switchover_status from v$database;
to primary
alter database commit to switchover to primary;
shutdown immediate;
startup
alter system switch logfile;
重复验证dg。
=================
注意：
一般switchover切换都是计划中的切换，特点是切换后，不会丢失任何数据，而且这个过程是
可逆的，整个data guard环境不会被破坏，原来的data guard环境中的所有standby
都可以继续工作。
在进行data guard的物理standby切换前：
1）确认主从库之间的网络连接畅通；
2）确认没有活动的会话连接在数据库中；
3）primary数据库处于打开状态，standby数据库处于mount状态。

来自 “ ITPUB博客 ” ，链接：http://blog.itpub.net/30592920/viewspace-2143889/，如需转载，请注明出处，否则将追究法律责任。

转载于:http://blog.itpub.net/30592920/viewspace-2143889/