一步一步超详细教你搭建Oracle19c ADG物理备库（Dupicate方式）

配置主库
开启主库归档模式
启动数据库到Mount模式
SYS@orcl>startup mount;
设置本地归档日志的存放路径
SYS@orcl>alter system set log_archive_dest_1='LOCATION=/oraarch/orcl';
修改数据库为归档模式
SYS@orcl>alter database archivelog;
打开强制日志模式
SYS@orcl>alter database force logging;
打开数据库
SYS@orcl>alter database open;
确认归档模式
SYS@orcl>archive log list;
Database log mode              Archive Mode
Automatic archival             Enabled
Archive destination            /oraarch/orcl
Oldest online log sequence     2
Next log sequence to archive   6
Current log sequence           6
确认强制日志模式
SYS@orcl>select log_mode,force_logging from v$database;

LOG_MODE     FORCE_LOGGING
------------ ---------------------------------------
ARCHIVELOG   YES

避坑提醒：如果DBCA建库时选择了打开归档模式并且开启了快速恢复区，请调大快速恢复区的容量限制，否则会发生磁盘没写满却无法归档日志的情况！

查询现有redo日志的数量
SYS@orcl>select group#,type,member from v$logfile;

    GROUP# TYPE    MEMBER
---------- ------- ------------------------------
         1 ONLINE  /oradata/ORCL/redo01.log
         2 ONLINE  /oradata/ORCL/redo02.log
         3 ONLINE  /oradata/ORCL/redo03.log
         4 ONLINE  /oradata/ORCL/redo04.log
         5 ONLINE  /oradata/ORCL/redo05.log


查询现有redo日志的容量
SYS@orcl>select group# ,bytes/1024/1024 as MB,status from v$log;

GROUP#   MB STATUS
---------- ---------- ----------------
1        256 CURRENT
2        256 INACTIVE
3        256 INACTIVE
4        256 INACTIVE
5        256 INACTIVE

给主库创建standby的redo日志组，方便角色切换为备库时使用，容量要和redo相同，数量是redo当前数量+1
alter database add standby logfile group 6 ('/oradata/ORCL/standbyredo06.log') size 256m;
alter database add standby logfile group 7 ('/oradata/ORCL/standbyredo07.log') size 256m;
alter database add standby logfile group 8 ('/oradata/ORCL/standbyredo08.log') size 256m;
alter database add standby logfile group 9 ('/oradata/ORCL/standbyredo09.log') size 256m;
alter database add standby logfile group 10 ('/oradata/ORCL/standbyredo10.log') size 256m;
alter database add standby logfile group 11 ('/oradata/ORCL/standbyredo11.log') size 256m;
确认standby redo logfile创建成功
SYS@orcl>select group#,status,used from v$standby_log;

    GROUP# STATUS           USED
---------- ---------- ----------
         6 UNASSIGNED          0
         7 UNASSIGNED          0
         8 UNASSIGNED          0
         9 UNASSIGNED          0
        10 UNASSIGNED          0
        11 UNASSIGNED          0

准备密码文件
[oracle@primary ~]$ cd $ORACLE_HOME/dbs/
[oracle@primary /oracle/app/product/19c/db_1/dbs]$ cp orapworcl /tmp/orapworclstd
设置主库的参数文件
SYS@orcl>create pfile from spfile;
[oracle@primary /oracle/app/product/19c/db_1/dbs]$ vim initorcl.ora
注意：备库内容最好在末行添加，最好不要使用alter system set来设置，后期不容易维护参数文件




#数据库唯一名称
*.db_unique_name='orcl'
*.log_archive_config='dg_config=(orcl,orclstd)'
#fal_client填写本机数据库唯一名称
*.fal_client='orcl' 
#fal_server填写对端的数据库唯一名称，多个备库时此处可以填写多个用逗号间隔
*.fal_server='orclstd'
#本地归档文件路径
*.log_archive_dest_1='LOCATION=/oraarch/orcl valid_for=(all_logfiles,all_roles) db_unique_name=orcl'
#远程归档文件传输给的服务名和数据库唯一名称
*.log_archive_dest_2='service=orclstd lgwr async valid_for=(online_logfiles,primary_role) db_unique_name=orclstd'
#建议归档日志改为arc结尾，避免被误删
*.log_archive_format='%t_%s_%r.arc'
#路径转换原则，本机的路径在后面，远程的路径放前面；路径映射应该成对出现，一个路径不能对应多个，要逐一进行映射；数据文件分散在多个路径时用逗号间隔；格式要严格一致，要么都加'/'，要么都不加
*.db_file_name_convert='/oradata/ORCLSTD/','/oradata/ORCL/','/oratemp/ORCLSTD/','/oratemp/ORCL/','/oraundo/ORCLSTD/','/oraundo/ORCL/'
#redo日志多路复写时记得增加映射路径
*.log_file_name_convert='/oradata/ORCLSTD/','/oradata/ORCL/'
STANDBY_FILE_MANAGEMENT=AUTO

启用新的参数文件
首先备份之前的spfile
SYS@orcl>create pfile='/home/oracle/spfile1028.bak' from spfile;
关闭实例
SYS@orcl>shutdown immediate
使用设置后的pfile替换当前spfile
SYS@orcl>create spfile from pfile;
启动实例
SYS@orcl>startup

传输参数文件和密码文件到备库
[oracle@primary /orabackup]$ cd /tmp/
[oracle@primary /tmp]$ scp orapworclstd oracle@standby:/tmp
[oracle@primary /tmp]$ cd $ORACLE_HOME/dbs
[oracle@primary /oracle/app/product/19c/db_1/dbs]$ scp initorcl.ora oracle@standby:/tmp/initorclstd.ora

设置备库
设置备库的SID
[oracle@standby ~]$  export ORACLE_SID=orclstd
将密码文件和参数文件拷贝拷贝到ORACLE的家目录
[oracle@standby ~]$ cd /tmp/
[oracle@standby /tmp]$ cp orapworclstd initorclstd.ora $ORACLE_HOME/dbs/
设置备库的参数文件
[oracle@standby /tmp]$ cd $ORACLE_HOME/dbs
[oracle@standby /oracle/app/product/19c/db_1/dbs]$ vim initorclstd.ora
#需要修改的部分
*.audit_file_dest='/oracle/app/admin/orclstd/adump'
*.control_files='/oradata/ORCLSTD/control01.ctl','/oradata/ORCLSTD/control02.ctl'

*.db_unique_name='orclstd'
*.log_archive_config='dg_config=(orcl,orclstd)'
*.fal_client='orclstd'
*.fal_server='orcl'
*.log_archive_dest_1='LOCATION=/oraarch/orclstd valid_for=(all_logfiles,all_roles) db_unique_name=orclstd'
*.log_archive_dest_2='service=orcl lgwr async valid_for=(online_logfiles,primary_role) db_unique_name=orcl'
*.log_archive_format='%t_%s_%r.arc'
*.db_file_name_convert='/oradata/ORCL/','/oradata/ORCLSTD/','/oratemp/ORCL/','/oratemp/ORCLSTD/','/oraundo/ORCL/','/oraundo/ORCLSTD/'
*.log_file_name_convert='/oradata/ORCL/','/oradata/ORCLSTD/'

创建备库相应的文件夹
mkdir -pv /oracle/app/admin/orclstd/adump
mkdir -pv /oradata/ORCLSTD
mkdir -pv /oraarch/orclstd
mkdir -pv /oratemp/ORCLSTD
mkdir -pv /oraundo/ORCLSTD
通过编辑的pfile启动备库的伪实例到nomount模式
[oracle@standby /]$ export ORACLE_SID=orclstd
[oracle@standby /]$ sqlplus / as sysdba
SYS@orclstd>create spfile from pfile;
SYS@orclstd>startup nomount;

配置网络
主库和备库做静态注册
配置主库的listener.ora文件
[oracle@primary ~]$ cd $ORACLE_HOME/network/admin/
[oracle@primary /oracle/app/product/19c/db_1/network/admin]$ vim listener.ora

LISTENER =
  (DESCRIPTION_LIST =
    (DESCRIPTION =
      (ADDRESS = (PROTOCOL = TCP)(HOST = primary)(PORT = 2521))
      (ADDRESS = (PROTOCOL = IPC)(KEY = EXTPROC2521))
    )
  )

SID_LIST_LISTENER =
  (SID_LIST =
    (SID_DESC =
     (GLOBAL_DBNAME = orcl)
     (ORACLE_HOME = /oracle/app/product/19c/db_1)
     (SID_NAME = orcl)
    )
   )

配置主库的tnsnames.ora
[oracle@primary /oracle/app/product/19c/db_1/network/admin]$ vim tnsnames.ora

LISTENER_ORCL =
  (ADDRESS = (PROTOCOL = TCP)(HOST = primary)(PORT = 2521))


ORCL =
  (DESCRIPTION =
    (ADDRESS = (PROTOCOL = TCP)(HOST = primary)(PORT = 2521))
    (CONNECT_DATA =
      (SERVER = DEDICATED)
      (SERVICE_NAME = orcl)
    )
  )

ORCLSTD =
  (DESCRIPTION =
    (ADDRESS = (PROTOCOL = TCP)(HOST = standby)(PORT = 2521))
    (CONNECT_DATA =
      (SERVER = DEDICATED)
      (SERVICE_NAME = orclstd)
    )
  )

配置备库的listener.ora文件
[oracle@standby ~]$ cd $ORACLE_HOME/network/admin
[oracle@standby /oracle/app/product/19c/db_1/network/admin]$ vim listener.ora

LISTENER =
  (DESCRIPTION_LIST =
    (DESCRIPTION =
      (ADDRESS = (PROTOCOL = TCP)(HOST = standby)(PORT = 2521))
      (ADDRESS = (PROTOCOL = IPC)(KEY = EXTPROC2521))
    )
  )

配置备库的tnsnames.ora
[oracle@standby /oracle/app/product/19c/db_1/network/admin]$ vim listener.ora

LISTENER_ORCLSTD =
  (ADDRESS = (PROTOCOL = TCP)(HOST = standby)(PORT = 2521))

ORCL =
  (DESCRIPTION =
    (ADDRESS = (PROTOCOL = TCP)(HOST = primary)(PORT = 2521))
    (CONNECT_DATA =
      (SERVER = DEDICATED)
      (SERVICE_NAME = orcl)
    )
  )

ORCLSTD =
  (DESCRIPTION =
    (ADDRESS = (PROTOCOL = TCP)(HOST = standby)(PORT = 2521))
    (CONNECT_DATA =
      (SERVER = DEDICATED)
      (SERVICE_NAME = orclstd)
    )
  )

重载主库和备库的监听器
[oracle@primary ~]$ lsnrctl reload
[oracle@standby ~]$ lsnrctl reload
静态监听注册成功之后使用lsnrctl stat查看监听器状态会显示如下内容
  Instance "orcl", status UNKNOWN, has 1 handler(s) for this service...
  Instance "orclstd", status UNKNOWN, has 1 handler(s) for this service...
网络连接测试
在主库和备库分别执行以下指令
tnsping orcl
tnsping orclstd
[oracle@primary ~]$ sqlplus sys/oracle@orclstd as sysdba
[oracle@standby ~]$ sqlplus sys/oracle@orcl as sysdba

使用Dupicate方式构建备库
[oracle@primary ~]$ rman target sys/oracle@orcl auxiliary sys/oracle@orclstd
注意以下的connected状态，主库显示DBID，备库为not mounted状态是正确的
Recovery Manager: Release 19.0.0.0.0 - Production on Fri Oct 29 10:02:47 2021
Version 19.10.0.0.0

Copyright (c) 1982, 2019, Oracle and/or its affiliates.  All rights reserved.

connected to target database: ORCL (DBID=1614658538)
connected to auxiliary database: ORCL (not mounted)

在RMAN中执行以下run块内容
run {
allocate channel ch001 type disk;
allocate channel ch002 type disk;
allocate channel ch003 type disk;
allocate channel ch004 type disk;
allocate auxiliary channel ch005 type disk;
allocate auxiliary channel ch006 type disk;
duplicate target database for standby from active database nofilenamecheck using compressed backupset;
release channel ch001;
release channel ch002;
release channel ch003;
release channel ch004;
release channel ch005;
release channel ch006;
}

备库初始化过程结束之后为MOUNT模式
SYS@orclstd>select open_mode from v$database;

OPEN_MODE
--------------------
MOUNTED

启用备库ADG模式
打开备库
SYS@orclstd>alter database open;
备库检查数据文件、redo log、standby redo log均生成：
set lines 222 pages 222
col name for a66
col member for a55
SYS@orclstd>select name from v$datafile;

NAME
------------------------------------------------------------------
/oradata/ORCLSTD/system01.dbf
/oradata/ORCLSTD/sysaux01.dbf
/oraundo/ORCLSTD/undotbs01.dbf
/oradata/ORCLSTD/appindex1.dbf
/oradata/ORCLSTD/appdata1.dbf
/oradata/ORCLSTD/appdata2.dbf
/oradata/ORCLSTD/appdata3.dbf

SYS@orclstd>select name from v$tempfile;

NAME
------------------------------------------------------------------
/oratemp/ORCLSTD/temp01.dbf
/oratemp/ORCLSTD/apptemp.dbf

SYS@orclstd>select GROUP#,TYPE,MEMBER from v$logfile order by 2,1;

    GROUP# TYPE    MEMBER
---------- ------- -------------------------------------------------------
         1 ONLINE  /oradata/ORCLSTD/redo01.log
         2 ONLINE  /oradata/ORCLSTD/redo02.log
         3 ONLINE  /oradata/ORCLSTD/redo03.log
         4 ONLINE  /oradata/ORCLSTD/redo04.log
         5 ONLINE  /oradata/ORCLSTD/redo05.log
         6 STANDBY /oradata/ORCLSTD/standbyredo06.log
         7 STANDBY /oradata/ORCLSTD/standbyredo07.log
         8 STANDBY /oradata/ORCLSTD/standbyredo08.log
         9 STANDBY /oradata/ORCLSTD/standbyredo09.log
        10 STANDBY /oradata/ORCLSTD/standbyredo10.log
        11 STANDBY /oradata/ORCLSTD/standbyredo11.log

开启日志应用
SYS@orclstd> alter database recover managed standby database using current logfile disconnect;

备库检查数据同步状态
col NAME for a23
col VALUE for a13
col TIME_COMPUTED for a20
col DATUM_TIME for a20
SYS@orclstd> select name,value,TIME_COMPUTED,DATUM_TIME from v$dataguard_stats;
该视图通过判断transport lag和apply lag两个参数的DATUM_TIME值来判断备库是否在应用主库的日志文件，如果多次查询DATUM_TIME值保持不变，则说明备注已经停止从主库接收数据了
NAME                    VALUE         TIME_COMPUTED        DATUM_TIME
----------------------- ------------- -------------------- --------------------
transport lag           +00 00:00:00  10/29/2021 10:59:44  10/29/2021 10:59:43
apply lag               +00 00:00:00  10/29/2021 10:59:44  10/29/2021 10:59:43
apply finish time       (null)        10/29/2021 10:59:44  (null)
estimated startup time  8             10/29/2021 10:59:44  (null)

检查MRP进程状态为APPLYING_LOG为正常同步状态
SYS@orclstd>select process,status,sequence# from v$managed_standby;

PROCESS     STATUS          SEQUENCE#
----------- --------------- ---------
RFS         IDLE                    0
ARCH        CONNECTED               0
DGRD        ALLOCATED               0
DGRD        ALLOCATED               0
ARCH        CONNECTED               0
ARCH        CONNECTED               0
ARCH        CONNECTED               0
RFS         IDLE                    9
RFS         IDLE                    0
MRP0        APPLYING_LOG            9



查看数据库当前打开状态为READ ONLY WITH APPLY则代表ADG启动成功
SYS@orclstd>select open_mode from v$database;

OPEN_MODE
--------------------
READ ONLY WITH APPLY