（一）基础环境
| 主库 | 备库 | |
| 操作系统 | RedHat6.7 | RedHat6.7 |
| 服务器名称 | primarydb1 primarydb2 |
standbydb1 standbydb2 |
| IP地址规划 |
192.168.10.31 primarydb1 10.10.10.31 primarydb1-priv |
192.168.10.41 standbydb1 10.10.10.41 standbydb1-priv |
| ------------------------ | ------------------------------ | --------------------------------------- |
| 数据库版本 | 11.2.0.4 | 11.2.0.4 |
| db_name | testdb | testdb |
| db_unique_name | testdb | testdbdg |
| instance_name | testdb1 testdb2 |
testdbdg1 testdbdg2 |
| service_name | testdb | testdbdg |
| 数据库安装情况 | 安装GI+数据库软件+创建数据库 | 安装GI+数据库软件 （不用创建数据库） |
在配置dataguard之前，主库已经安装了GI+数据库软件+创建了数据库，备库也已经安装了GI+数据库软件，备库不需要使用dbca建库。
（二）主库配置
（2.1）配置归档，主库需要运行在归档模式下
--查看是否开启归档 archive log list
如果没有开启归档，需要开启，方法如下：
step1: 设置归档日志文件存放地址
alter system set log_archive_dest_1='LOCATION=+ARCH' ;
step2: 关闭数据库，2个节点都关闭
shutdown immediate
step3: 重启节点1到mount状态
startup mount
step4: 开启归档
alter syetem archivelog
step5: 打开数据库
alter database open
step6: 确认归档状态
archive log list
最好使用alter system switch logfile切换日志，确认日志已经存到归档位置
step7：打开另外一个节点
startup open;
（2.2）主库启用强制记录日志
(2.3)主库网络配置
（2.3.1）静态监听配置
主库可以使用动态监听，也可以使用静态监听，这里新创建一个监听LISTENER_ADG，监听新的端口1522。以下操作使用grid用户执行
step1:在节点1执行添加监听命令
srvctl add listener -l LISTENER_ADG -o $ORACLE_HOME -p "TCP:1522"
step2:启动监听
srvctl start listener -l LISTENER_ADG
step3:修改监听配置文件，加入静态注册信息，2个节点都需要修改，这里以节点1为例
[grid@primarydb1 ~]$ vim /u01/app/11.2.0/grid/network/admin/listener.ora
LISTENER_ADG=(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=IPC)(KEY=LISTENER_ADG)))) # line added by Agent LISTENER=(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=IPC)(KEY=LISTENER)))) # line added by Agent LISTENER_SCAN1=(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=IPC)(KEY=LISTENER_SCAN1)))) # line added by Agent ENABLE_GLOBAL_DYNAMIC_ENDPOINT_LISTENER_SCAN1=ON # line added by Agent ENABLE_GLOBAL_DYNAMIC_ENDPOINT_LISTENER=ON # line added by Agent ENABLE_GLOBAL_DYNAMIC_ENDPOINT_LISTENER_ADG=ON # line added by Agent #添加以下信息 SID_LIST_LISTENER_ADG = (SID_LIST = (SID_DESC = (GLOBAL_DBNAME = testdb) (ORACLE_HOME = /u01/app/11.2.0/grid) #填写的是grid_home (SID_NAME=testdb1) #如果是节点2，需改为SID_NAME=testdb2 ) )
step4：重启监听，一个节点执行
srvctl stop listener -l LISTENER_ADG
srvctl start listener -l LISTENER_ADG
（2.3.2）tns配置，主库2个节点都添加
[oracle@primarydb1 ~]$ cd /u01/app/oracle/product/0/db_1/network/admin/ [oracle@primarydb1 admin]$ ls samples shrept.lst tnsnames.ora [oracle@primarydb1 admin]$ vim tnsnames.ora
tnstestdb = (DESCRIPTION = (ADDRESS = (PROTOCOL = TCP)(HOST = 192.168.10.33)(PORT = 1522)) (ADDRESS = (PROTOCOL = TCP)(HOST = 10.34)(PORT = ))
(CONNECT_DATA = (SERVER = DEDICATED) (SERVICE_NAME = testdb) ) ) tnstestdbdg = (DESCRIPTION =10.43)(PORT = ))
(ADDRESS = (PROTOCOL = TCP)(HOST = 10.44)(PORT = )) (CONNECT_DATA = testdbdg) ) )
(2.4)主库参数配置
set db_unique_name = testdb' scope=spfile;
set log_archive_configDG_CONFIG=(testdb,testdbdg);
set log_archive_dest_2SERVICE=tnstestdbdg LGWR SYNC AFFIRM VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=testdbdgset log_archive_dest_state_1 "enable";
set log_archive_dest_state_2 set db_file_name_convert+DATA/testdbdg',+DATA/testdbset log_file_name_convertset standby_file_managementauto;
set fal_clienttnstestdbset fal_servertnstestdbdg';
因为有的参数重启才会生效（scope=spfile），因此改完参数需要重启
[grid@primarydb1 ~]$ srvctl stop database -d testdb
[grid@primarydb1 ~]$ srvctl start database -d testdb
(2.5)添加standby online redo log
standby redo log的大小与redo log大小相同。组数为比在线日志多一组。
查看在线日志组数信息
sql> select a."GROUP#",a."THREAD#",a."BYTES",a."MEMBERS" from v$ a,v$logfile b where a."GROUP#" = b."#" order by group#; # THREAD# BYTES MEMBERS -------- ---------- ---------- ---------- 1 1 52428800 1 2 3 2 4 1
可以看到，每个线程（可以理解为每个实例）有2组日志文件，每组有1个日志文件。因此我们在创建standby redo log时需要为每个thread创建3组日志文件。
thread1
add standby logfile thread 1 group 11 (+DATA) size 50M;
12 (13 (thread2
2 14 (15 (16 (') size 50M;
（2.6）将主库的密码文件拷贝到备库
拷贝主库密码文件到备库，在节点1执行
[oracle@primarydb1 dbs]$ pwd /u01/app/oracle/product/0/db_1/dbs # 拷贝密码文件到备库节点1 [oracle@primarydb1 dbs]$ scp orapwtestdb1 oracle@10.41:`pwd` # 拷贝密码文件到备库节点2 [oracle@primarydb1 dbs]$ 10.42:`pwd`
（2.7）将主库的参数文件拷贝到备库
sql> create pfile='/home/oracle/pfile_20190818 from spfile; sql> exit Disconnected from Oracle Database 11g Enterprise Edition Release 0.4.0 - 64bit Production With the Partitioning,Real Application Clusters,Automatic Storage Management,OLAP,Data Mining and Real Application Testing options [oracle@primarydb1 ~]$ scp pfile_20190818 oracle@` oracle@10.41s password: pfile_20190818 100% 1859 1.8KB/s 00:00
（三）备库配置
linux环境中，密码文件命名格式为：orapwd{$sid}，因此需要修改密码文件的名字
备库节点1：
[oracle@standbydb1 dbs]$ dbs [oracle@standbydb1 dbs]$ mv orapwtestdb1 orapwtestdbdg1
备库节点2：
[oracle@standbydb2 dbs]$ dbs [oracle@standbydb2 dbs]$ mv orapwtestdb1 orapwtestdbdg2
最终参数文件如下：
*.audit_file_dest=/u01/app/oracle/admin/testdbdg/adump' *.audit_trail=db' *.cluster_database=true *.compatible=11.2.0.4.0' *.control_files=+DATA/testdb/controlfile/current.266.1016639457' *.db_block_size=8192 *.db_create_file_dest=+DATA' *.db_domain='' *.db_file_name_convert=+DATA/testdb+DATA/testdbdg' *.db_name=testdb' *.db_unique_name=testdbdg' *.diagnostic_dest=/u01/app/oracle' *.fal_client=tnstestdbdg' *.fal_server=tnstestdb testdbdg1.instance_number=1 testdbdg2.instance_number=2 *.log_archive_config=' *.log_archive_dest_1=LOCATION=+ARCH' *.log_archive_dest_2=SERVICE=tnstestdb LGWR SYNC AFFIRM VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=testdb' *.log_archive_dest_state_1=enable' *.log_archive_dest_state_2=' *.log_file_name_convert=' *.remote_login_passwordfile=exclusive' *.standby_file_management=AUTO testdb1.thread= testdb2.thread=2 testdb1.undo_tablespace=UNDOTBS2 testdb2.undo_tablespace=UNDOTBS1'
里面涉及到的路径需要手动创建
su - oracle
# 2个节点都要创建
mkdir -p mkidr -p /u01/app/oracle/admin/testdbdg/adump
（3.3）备库网络配置
（3.3.1）静态监听配置
这里新创建一个监听LISTENER_ADG，监听新的端口1522。以下操作使用grid用户执行
step1:在节点1执行添加监听命令
srvctl add listener -l LISTENER_ADG -o $ORACLE_HOME -p "TCP:1522"
step2:启动监听
srvctl start listener -l LISTENER_ADG
step3:修改监听配置文件，加入静态注册信息，2个节点都需要修改，这里以节点1为例
[grid@standbydb1 ~]$ vim /u01/app/0/grid/network/admin/listener.ora LISTENER_ADG=(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=IPC)(KEY=ON # line added by Agent SID_LIST_LISTENER_ADG = testdbdg) (ORACLE_HOME = /u01/app/grid) (SID_NAME=testdbdg1) ) )
step4：重启监听，一个节点执行
l LISTENER_ADG
srvctl start listener -l LISTENER_ADG
（3.3.2）tns配置，备库2个节点都添加
samples shrept.lst tnsnames.ora
[oracle@primarydb1 admin]$ vim tnsnames.ora
# 在tnsnames.ora文件中添加如下内容
tnstestdb = testdbdg)
)
)
(3.4)启动备库到nomount状态
[oracle@standbydb1 ~]$ sqlplus / as sysdba sql*Plus: Release 0 Production on Mon Aug 19 04:14:04 2019 Copyright (c) 1982,1)">2013,Oracle. All rights reserved. Connected to an idle instance. sql> startup nomount pfile='; ORACLE instance started. Total System Global Area 313196544 bytes Fixed Size 2252744 bytes Variable Size 255852600 bytes Database Buffers 50331648 bytes Redo Buffers 4759552 bytes sql>
注意：再将备库启动到nomount状态后，理论上，可以从主库通过sqlplus连接到备库，同样也可以在备库通过sqlplus连接到主库，这里建议测试主库备库之间网络的连通性。在主库与备库上执行
确保所有命令都能执行成功。
在节点1的oracle用户下执行：
[oracle@standbydb1 db_1]$ srvctl add database -d testdbdg -o /u01/app/oracle/product/db_1 [oracle@standbydb1 db_1]$ srvctl add instance -d testdbdg -i testdbdg1 -n standbydb1 [oracle@standbydb1 db_1]$ srvctl add instance -d testdbdg -i testdbdg2 -n standbydb2
（四）主库备份
rman target / RMAN> run { allocate channel c1 type disk; allocate channel c2 type disk; sql alter system archive log current; backup database format /databaseBackup/full_db_%U; sql; backup archivelog all format /databaseBackup/archlog_%U; backup current controlfile format /databaseBackup/controlfile_%U; backup spfile format /databaseBackup/spfile_%U; release channel c1; release channel c2; }
将备份传到备库服务器：
[oracle@primarydb1 /]$ cd databaseBackup/ [oracle@primarydb1 databaseBackup]$ archlog_0au9imjq_1_1 archlog_0cu9imnk_1_1 full_db_06u9imbr_1_1 full_db_08u9imhs_1_1 spfile_0gu9imtt_1_1 archlog_0bu9imju_1_1 controlfile_0du9imrd_1_1 full_db_07u9imbr_1_1 full_db_09u9imhv_1_1 [oracle@primarydb1 databaseBackup]$ scp * oracle@10.41:/databaseBackup/
（五）备库还原
（5.1）restore控制文件
[oracle@standbydb1 ~]$ rman target / Recovery Manager: Release 0 - Production on Mon Aug 06:47:00 2011,Oracle and/or its affiliates. All rights reserved. connected to target database: TESTDB (not mounted) RMAN> restore standby controlfile from /databaseBackup/controlfile_0du9imrd_1_1'; Starting restore at 19-AUG-19 using target database control file instead of recovery catalog allocated channel: ORA_DISK_1 channel ORA_DISK_1: SID=292 instance=testdbdg1 device type=DISK channel ORA_DISK_1: restoring control channel ORA_DISK_1: restore complete,elapsed time: 08 output file name=+DATA/testdbdg/controlfile/current.268.1016664445 Finished restore at 19
解决方法：https://jingyan.baidu.com/album/9faa7231c13269473c28cb33.html?picindex=9
（5.2）修复数据库
在启动到mount状态时报了2个错误：
报错1：初始化参数文件找不到
解决方法：复制pfile到给定位置
[oracle@standbydb1 bin]$ cp /home/oracle/pfile_20190818 /u01/app/oracle/product/11.2.0/db_1/dbs/inittestdbdg1.ora
报错2：控制文件找不到
解决方法：
1.通过asmcmd找到控制文件的位置；
ASMCMD> pwd +data/testdbdg/CONTROLFILE ASMCMD> current.256.1016666437
[oracle@standbydb1 ~]$ vim /u01/app/oracle/product/0/db_1/dbs/inittestdbdg1.ora ... *.control_files='+data/testdbdg/CONTROLFILE/current.256.1016666437'
...
重新启动数据库到mount状态
[oracle@standbydb1 dbs]$ sqlplus /07:08:15 sql> startup mount; ORACLE instance started. Total System Global Area bytes Database mounted.
开始修复数据库
RMAN> restore database; Starting restore at 14 instance=testdbdg1 device type=DISK channel ORA_DISK_1: starting datafile backup set restore channel ORA_DISK_1: specifying datafile(s) to restore from backup set channel ORA_DISK_1: restoring datafile 00001 to +DATA/testdbdg/datafile/system.268.1016639305 channel ORA_DISK_1: restoring datafile 00004 to +DATA/testdbdg/datafile/users.267.1016639305 channel ORA_DISK_1: reading from backup piece /databaseBackup/full_db_06u9imbr_1_1 channel ORA_DISK_1: piece handle=/databaseBackup/full_db_06u9imbr_1_1 tag=TAG20190819T035434 channel ORA_DISK_1: restored backup piece channel ORA_DISK_1: starting datafile backup set restore channel ORA_DISK_1: specifying datafile(s) to restore from backup set channel ORA_DISK_1: restoring datafile 00002 to +DATA/testdbdg/datafile/sysaux.260.101663930500003 to +DATA/testdbdg/datafile/undotbs1.263.101663930500005 to +DATA/testdbdg/datafile/undotbs2.264.1016639859full_db_07u9imbr_1_1 channel ORA_DISK_1: piece handle=/databaseBackup/full_db_07u9imbr_1_1 tag=5619
（六）打开备库
打开备库
sql> alter database open ; Database altered.
查看数据库状态：
sql> select name,db_unique_name,open_mode from v$database; NAME DB_UNIQUE_NAME OPEN_MODE --------- ------------------------------ -------------------- TESTDB testdbdg READ ONLY
开始数据库实时日志应用,在开启之前，需要确认standby redo logfile是否已经创建，这里已经从主库恢复过来，故不需要创建
database recover managed standby database using current logfile disconnect;
Database altered.
在执行该语句后，数据库会自动去同步主库新生成的日志，可以打开主库与备库的告警日志，查看数据库同步状态。
（七）启动另外一个节点
（7.1）首先使用pfile生成spfile
create spfile+DATA/testdbdg/spfiletestdbdg.ora' from pfile/u01/app/oracle/product/11.2.0/db_1/dbs/inittestdbdg1.ora;
File created.
shutdown immediate
Database closed.
Database dismounted.
ORACLE instance shut down.
#节点1
[oracle@standbydb1 ~]$ cd /u01/app/oracle/product/
[oracle@standbydb1 dbs]$ vim inittestdbdg1.ora
# 仅保留一行参数，指向spfile
spfile=+DATA/testdbdg/spfiletestdbdg.ora
#节点2
[oracle@standbydb1 ~]$ cd /u01/app/oracle/product/
[oracle@standbydb1 dbs]$ vim inittestdbdg2.ora
# 仅保留一行参数，指向spfile
spfile='
（7.2）启动节点
首先查看数据库状态：
[grid@standbydb1 ~]$ crsctl status res -t -------------------------------------------------------------------------------- ... ora.testdbdg.db OFFLINE OFFLINE Instance Shutdown OFFLINE OFFLINE ...
打开节点1：
打开节点2：
oracle@standbydb2 dbs*Plus: Release 11.2.0.4.0 Production on Thu Aug 22 06:07:55
Copyright (c) 2013,Oracle. All rights reserved.
Connected to an idle instance.
sql startup
ORACLE instance started.
Total System Global Area bytes
Fixed Size bytes
Variable Size bytes
Database Buffers bytes
Redo Buffers Database mounted.
opened.
再次确认数据库的状态,已经正常启动：
grid@standbydb1 ~]$ crsctl status res -t
------------------------------------------------------------------------------
...
ora.testdbdg.db
1 ONLINE ONLINE standbydb1 Open2 ONLINE ONLINE standbydb2 数据库状态：
主库：
select name,open_mode,database_role ;
NAME DB_UNIQUE_NAME OPEN_MODE DATABASE_ROLE
------- ------------------------------ -------------------- ----------------
TESTDB testdb READ WRITE PRIMARY
备库：
------- ------------------------------ -------------------- ----------------
TESTDB testdbdg READ ONLY WITH APPLY PHYSICAL STANDBY
（八）测试数据同步情况
（1）主库创建表，插入数据：
create table test02 (id number,name varchar(20));
Table created.
sqlinsert into test02 values(1,1)">lijiaman);
row created.
sqlcommitCommit complete.
sql> select * from test02;
ID NAME
---------- --------------------
1 lijiaman
备库查看：
select * from test02;
ID NAME
-------- --------------------
1 lijiaman
数据已同步。
（2）主库删除表，表进入了回收站，11gR2的dataguard不需要关闭回收站。
drop table test02;
dropped.
sql tab;
TNAME TABTYPE CLUSTERID
---------------------------- ------- ----------
BIN$kKkJUDt4hvrgUyAKqMC61w==$0 TABLE
发现备库表也进入了回收站
TABLE
同样，主库执行闪回删除，将sales表从回收站恢复回来，备库也会执行相同的操作。
> flashback table test02 to before drop;
Flashback complete.
sql tab;
TNAME TABTYPE CLUSTERID
---------------------------- ------- ----------
TEST02 TABLE
TABLE
数据同步测试无异常。
（九）存在的问题
存在的问题1：备库使用scan-ip无法连接：
配置信息：ip192.168.10.40 port:1521 service:testdbdg
需要设置初始化参数：remote_listener
> set remote_listenerstandbydb-scan:1521'
【完】
