19c_dg

1主库

ALTER DATABASE FORCE LOGGING;
ALTER DATABASE SET STANDBY NOLOGGING FOR DATA AVAILABILITY;
ALTER DATABASE SET STANDBY NOLOGGING FOR LOAD PERFORMANCE;

1.1监听配置

ps:主库为asm单实例
[grid@ora2 admin]$ cat listener.ora
#Backup file is  /u01/app/grid/crsdata/ora2/output/listener.ora.bak.ora2.grid line added by Agent
SID_LIST_LISTENER = 
  (SID_LIST = 
    (SID_DESC = 
      (GLOBAL_DBNAME = orcl) 
      (ORACLE_HOME = /u01/app/oracle/product/19.0.0/db_1) 
      (SID_NAME = orcl) 
    ) 
  ) 
 
LISTENER = 
  (DESCRIPTION_LIST = 
    (DESCRIPTION = 
      (ADDRESS = (PROTOCOL = IPC)(KEY = LISTENER)) 
    ) 
    (DESCRIPTION = 
      (ADDRESS = (PROTOCOL = TCP)(HOST = 192.168.56.121)(PORT = 1521)) 
    ) 
  ) 
ENABLE_GLOBAL_DYNAMIC_ENDPOINT_LISTENER=ON		# line added by Agent
VALID_NODE_CHECKING_REGISTRATION_LISTENER=ON		# line added by Agent
[grid@ora2 admin]$ 

1.2主库tns

[oracle@ora2 admin]$ cat tnsnames.ora 
# Generated by Oracle configuration tools. PDBDGORCL = 
STD_ORCL = 
  (DESCRIPTION = 
    (ADDRESS = (PROTOCOL = TCP)(HOST = 192.168.56.120)(PORT = 1521)) 
    (CONNECT_DATA = 
      (SERVER = DEDICATED) 
      (SERVICE_NAME = orcl) 
    ) 
  ) 
 
ORCL = 
  (DESCRIPTION = 
    (ADDRESS = (PROTOCOL = TCP)(HOST = 192.168.56.121)(PORT = 1521)) 
    (CONNECT_DATA = 
      (SERVER = DEDICATED) 
      (SERVICE_NAME = orcl) 
    ) 
  ) 

1.3主库参数文件

[oracle@ora2 ~]$ cat pfile 
*.audit_file_dest='/u01/app/oracle/admin/orcl/adump'
*.audit_trail='db'
*.common_user_prefix=''
*.compatible='19.0.0'
*.control_files='+B/ORCL/CONTROLFILE/current.260.1139638345'
*.db_create_file_dest='+B'
*.db_name='orcl'
*.db_recovery_file_dest_size=5368709120
*.db_recovery_file_dest='/u01/flash'
*.dg_broker_start=TRUE
*.diagnostic_dest='/u01/app/oracle'
*.enable_pluggable_database=true
*.fal_server='STD_ORCL'
*.local_listener=''
*.log_archive_config='DG_CONFIG=(orcl,std_orcl)'
*.log_archive_dest_1='location=/home/oracle/arch VALID_FOR=(ALL_LOGFILES,ALL_ROLES) DB_UNIQUE_NAME=orc
l'
*.log_archive_dest_2='service=std_orcl async VALID_FOR=(ONLINE_LOGFILE,PRIMARY_ROLE) DB_UNIQUE_NAME=st
d_orcl'
*.log_archive_dest_state_1='enable'
*.log_archive_dest_state_2='enable'
*.log_file_name_convert='/u01/app/oracle/oradata/','+B/ORCL/ONLINELOG/'
*.db_file_name_convert='/u01/app/oracle/oradata','+B'
*.standby_file_management='AUTO'
*.temp_undo_enabled=TRUE
*.timed_statistics=TRUE

2.备库

2.1备库监听

[oracle@ora1 admin]$ cat listener.ora 
SID_LIST_LISTENER = 
  (SID_LIST = 
    (SID_DESC = 
      (GLOBAL_DBNAME = orcl) 
      (ORACLE_HOME = /u01/app/oracle/product/19.0.0/db_1) 
      (SID_NAME = orcl) 
    ) 
  ) 
 
LISTENER = 
  (DESCRIPTION_LIST = 
    (DESCRIPTION = 
      (ADDRESS = (PROTOCOL = IPC)(KEY = LISTENER)) 
    ) 
    (DESCRIPTION = 
      (ADDRESS = (PROTOCOL = TCP)(HOST = 192.168.56.120)(PORT = 1521)) 
    ) 
  ) 

2.2备库tns

从主库复制过来
[oracle@ora1 admin]$ cat tnsnames.ora 
STD_ORCL = 
  (DESCRIPTION = 
    (ADDRESS = (PROTOCOL = TCP)(HOST = 192.168.56.120)(PORT = 1521)) 
    (CONNECT_DATA = 
      (SERVER = DEDICATED) 
      (SERVICE_NAME = orcl) 
    ) 
  ) 
 
ORCL = 
  (DESCRIPTION = 
    (ADDRESS = (PROTOCOL = TCP)(HOST = 192.168.56.121)(PORT = 1521)) 
    (CONNECT_DATA = 
      (SERVER = DEDICATED) 
      (SERVICE_NAME = orcl) 
    ) 
  ) 

2.3备库pfile

*.control_files='/u01/app/oracle/oradata/control01.ctl'#Restore Controlfile
*.db_block_size=8192
*.db_create_file_dest='/u01/app/oracle/oradata'
*.db_name='orcl'
*.db_recovery_file_dest_size=5368709120
*.db_recovery_file_dest='/u01/flash'
*.db_unique_name='std_orcl'
*.dg_broker_start=TRUE
*.diagnostic_dest='/u01/app/oracle'
*.dispatchers='(PROTOCOL=TCP) (SERVICE=orclXDB)'
*.enable_pluggable_database=true
*.fal_client='STD_ORCL'
*.fal_server='orcl'
*.local_listener=''
*.log_archive_config='DG_CONFIG=(std_orcl,orcl)'
*.log_archive_dest_1='location=/home/oracle/arch valid_for=(all_logfiles,all_roles) db_unique_name=std
_orcl'
*.log_archive_dest_2='service=orcl async VALID_FOR=(ONLINE_LOGFILE,PRIMARY_ROLE) DB_UNIQUE_NAME=orcl'
*.log_file_name_convert='+B/ORCL/ONLINELOG/','/u01/app/oracle/oradata/'
*.db_file_name_convert='+B','/u01/app/oracle/oradata'
*.standby_file_management='AUTO'
*.temp_undo_enabled=TRUE
*.timed_statistics=TRUE

3.参数说明

1.DB_NAME
只需注意DataGuard的主备各节点instance使用相同的db_name即可。推荐与service_name一致。我们这里未作重定义
2.DB_UNIQUE_NAME
1）Primary与Standby端数据库的唯一名字，设定后不可再更改。
2）如果主备db_unique_name不一样，需要与LOG_ARCHIVE_CONFIG配合使用 
3）db_unique_name并未规定需要与数据库service_name一致，可以自定义任意名称。
3.LOG_ARCHIVE_CONFIG
列出主备库上的DB_UNIQUE_NAME 参数。默认情况下，定义该参数保证数据库能够发送或接收redo log。
4.LOG_ARCHIVE_DEST_1
本地归档路径。Primary与Standby需要定义各自的online redo log的归档地址，以系统实际的存放路径为准。
本例LOG_ARCHIVE_DEST_1='LOCATION=/u01/arch/ VALID_FOR=(ALL_LOGFILES,ALL_ROLES) DB_UNIQUE_NAME=bj'
可以这样理解：对于bj（主库）而言，不管它是主库角色还是备库角色（ALL_ROLES），它都会自己完成归档动作，并将它们归档在本地的/u01/arch下。
5.LOG_ARCHIVE_DEST_2
该参数仅当数据库角色为primary时生效，指定primary传输redo log到该参数定义的standby database上。
log_archive_dest_2可以说是dataguard上最重要的参数之一，它定义了redo log的传输方式(sync or async)以及传输目标(即standby apply node)，直接决定了dataguard的数据保护级别。
注意：SERVICE值，比如bj，是tnsnames.ora文件里定义的Oracle Net名称。
6.LOG_ARCHIVE_DEST_STATE_n
设置为ENABLE，激活log_archive_dest_n定义的属性。
7.FAL_SERVER and FAL_CLIENT
FAL是Fetch Archive Log的简写，它是dataguard主备之间GAP（间隙）的处理机制。
Primary上不会有GAP，所以fal_server和fal_client也是只在standby上生效的参数，当然为了switch over的需要同样会在primary端进行预设置。
FAL参数定义的数据库名同样取自本地tnsnames.ora里配置的Oracle Net Service Name.
11g以后 FAL_CLIENT可以忽略。
本例FAL_SERVER=tj的意思可以理解为:tj（备库）产生的归档日志间隙，要从主库（bj）那里获得。(fal_server=primary database db_unique_name)
8.DB_FILE_NAME_CONVERT
primary与standby上diskgroup的名称或是数据文件的存放路径不一致的时候，需要定义该参数进行转换，否则standby apply后无法创建与primary一致的数据文件并报错。
1）primary上的该参数仅在主备switch over后生效 
2）注意格式应保持一致，比如"*.db_file_name_convert='+DG1/db/datafile','+DATAGRP/db/datafile/' ”，路径少了一个"/”，将导致standby apply失败。
本例 DB_FILE_NAME_CONVERT='/u01/oradata/prodstd','/u01/oradata/prod' 使用的是物理文件路径，所以可以省略”/”
3）primary上执行create tablespace等add datafile操作时，无须自定义datafile的全路径名称，由数据库自动创建datafile即可。
9.LOG_FILE_NAME_CONVERT
同DB_FILE_NAME_CONVERT类似，定义主备log文件的存放路径转换。
10）STANDBY_FILE_MANAGEMENT
设置为auto时，主库执行增删数据文件的动作，在备库也会增删数据文件，缺省是manual 需要在备库做手动增删。

4.开始搭建

4.1 主库 添加standby log

1.查看redo log大小
SQL>  select bytes/1024/1024 M from v$log;

	 M
----------
       200
       200
       200
       
2.standby log要和 redolog  大小一致，否则后面mrp0进程状态一致waitapply状态 
alter database add standby  logfile group 4 '/u01/app/oracle/oradata/std1.log' size 200m;
alter database add standby  logfile group 5 '/u01/app/oracle/oradata/std2.log' size 200m;
alter database add standby  logfile group 6 '/u01/app/oracle/oradata/std3.log' size 200m;
alter database add standby  logfile group 7 '/u01/app/oracle/oradata/std4.log' size 200m;

3.若添加失败，删除重新添加
ALTER DATABASE DROP STANDBY LOGFILE GROUP 4;
ALTER DATABASE DROP STANDBY LOGFILE GROUP 5;
ALTER DATABASE DROP STANDBY LOGFILE GROUP 6;
ALTER DATABASE DROP STANDBY LOGFILE GROUP 7;

4.若添加standby log成功，查询 standby log大小及状态
SELECT v$logfile.member,
       v$logfile.group#,
       v$standby_log.status,
       v$standby_log.ARCHIVED,
       v$standby_log.bytes / 1024 / 1024 M,
       v$standby_log.thread#
  FROM v$standby_log, v$logfile
 WHERE v$standby_log.group# = v$logfile.group#
 ORDER BY v$standby_log.thread#, v$logfile.group#;

4.2启动备库到nomount状态，开始复制

[oracle@ora1 ~]$  rman target sys/oracle@orcl auxiliary sys/oracle@std_orcl

Recovery Manager: Release 19.0.0.0.0 - Production on Sat Jul 22 15:40:30 2023
Version 19.3.0.0.0

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

connected to target database: ORCL (DBID=1667419081)
connected to auxiliary database: ORCL (started)

RMAN> duplicate target database for standby from active database;

5.查看备库状态

5.1复制 进程控制

1.复制完成备库处于mount状态 ，可启动至open
2，启动mrp0进程
ALTER DATABASE RECOVER MANAGED STANDBY DATABASE DISCONNECT FROM SESSION;
ALTER DATABASE RECOVER MANAGED STANDBY DATABASE cancel;

5.2查看复制状态

SQL> select process,status from v$managed_standby;

PROCESS   STATUS
--------- ------------
DGRD	  ALLOCATED
ARCH	  CLOSING
DGRD	  ALLOCATED
ARCH	  CLOSING
ARCH	  CLOSING
ARCH	  CLOSING
MRP0	  APPLYING_LOG
RFS	  IDLE
RFS	  IDLE
RFS	  IDLE
RFS	  IDLE


SQL> SELECT ROLE, THREAD#, SEQUENCE#, ACTION FROM V$DATAGUARD_PROCESS;

ROLE			    THREAD#  SEQUENCE# ACTION
------------------------ ---------- ---------- ------------
log writer			  0	     0 IDLE
redo transport monitor		  0	     0 IDLE
gap manager			  0	     0 IDLE
redo transport timer		  0	     0 IDLE
archive local			  0	     0 IDLE
archive redo			  0	     0 IDLE
archive redo			  0	     0 IDLE
archive redo			  0	     0 IDLE
broker instance slave		  0	     0 IDLE
broker monitor			  0	     0 IDLE
managed recovery		  0	     0 IDLE
recovery logmerger		  1	   192 APPLYING_LOG
recovery apply slave		  0	     0 IDLE
recovery apply slave		  0	     0 IDLE
RFS archive			  0	     0 IDLE
RFS async			  1	   192 IDLE
RFS ping			  1	   192 IDLE
RFS archive			  0	     0 IDLE
RFS archive			  0	     0 IDLE

19 rows selected.

创建同步pdb

 create pluggable database pdbprod3 admin user pdbadmin identified by oracle;    

6.排错

如状态异常，可直接用以下 命令在sqlplus中修改配置
alter system set LOG_ARCHIVE_CONFIG='DG_CONFIG=(std_orcl,orcl)';
alter system set log_archive_dest_2='service=orcl async VALID_FOR=(ONLINE_LOGFILE,PRIMARY_ROLE) DB_UNIQUE_NAME=orcl';

6.1查看备库归档日志gap状态

select  *  from  v$archive_gap;
select  *  from  gv$archive_gap;

--查看备库归档应用情况：
set  linesize  600  pagesize  100
select thread#,max(sequence#) from v$log_history group by thread#;

---备库上查看同步过来的归档日志的应用情况：
col name for a80
select name,SEQUENCE#,APPLIED from v$archived_log order by sequence#;