现有用户数据库运行在超融合单机系统情况下，前端时间由于异常重启导致数据库system01.dbf文件损坏，核心数据库业务无法运行，本次考虑在超融合环境下搭建双机，为现有的数据库提供冗余环境，由于现有的核心业务运行在数据库上，停机时间控制在30分钟内（数据量2T），否则影响到业务系统的正常运行。
方案步骤安排如下：
1、新建双机集群环境，包括数据库，测试双机环境的可用性和可靠性；
2、停用集群，卸载新集群数据库实例，新老数据库搭建oracle adg，同步老库数据到新库；
3、搭建双机集群；
4、确认割接时间点，主备库切换；
5、带入双机集群，切换生产ip地址，测试双机切换；
6、取消新库adg参数，割接完成；
7、检查新集群运行状态。
备注：前3步搭建过程不影响业务正常运行，从第4步开始需要沟通具体割接时间。
一、情况说明
1、操作系统windows 2019标准版（本次需要升级操作系统从win2012到win2019）
2、数据库版本：11.2.0.4 for windows
3、地址信息
（1）源库（主）：192.168.10.100
（2）备库（集群）：192.168.10.200、192.168.10.201（ha集群另外一台）
4、源库windows系统oracle数据磁盘分多个存储空间
C:120g D:1.3t E:2.5t F:2t
目标库：C:120g D:3T E:2T
二、主库文件配置
1、启用强制日志模式
SQL> select force_logging from v$database;
2、配置重做认证（将密码文件复制到备机）
3、配置主数据库接收重做数据
4、参数初始化
alter system set LOG_ARCHIVE_CONFIG='DG_CONFIG=(orcl,orcldg)' sid='*' scope=both;
alter system set LOG_ARCHIVE_DEST_1='LOCATION=USE_DB_RECOVERY_FILE_DEST VALID_FOR=(ALL_LOGFILES,ALL_ROLES) DB_UNIQUE_NAME=orcl' sid='*' scope=both;
alter system set LOG_ARCHIVE_DEST_2='SERVICE=orcldg ASYNC VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=orcldg' sid='*' scope=both;
alter system set FAL_SERVER=orcldg sid='*' scope=both;
alter system set DB_FILE_NAME_CONVERT='d:\oradata\orcldg','d:\oradata\orcl' sid='*' scope=spfile;
alter system set LOG_FILE_NAME_CONVERT='d:\oradata\orcldg','d:\oradata\orcl' sid='*' scope=spfile;
alter system set STANDBY_FILE_MANAGEMENT=AUTO sid='*' scope=both;
5、配置tnsnames.ora文件
tnsnames.ora文件添加如下：
orcl =
(DESCRIPTION =
(ADDRESS = (PROTOCOL = TCP)(HOST = 192.168.10.100)(PORT = 1521))
(CONNECT_DATA =
(SERVER = DEDICATED)
(SERVICE_NAME = orcl)
)
)
orcldg =
(DESCRIPTION =
(ADDRESS = (PROTOCOL = TCP)(HOST = 192.168.10.200)(PORT = 1521))
(CONNECT_DATA =
(SERVER = DEDICATED)
(SERVICE_NAME = orcl)
)
)
备注：现有生产环境listener.ora不动，防止影响业务使用。
三、配置备库
1、备库安装数据库软件（过程略）
2、修改参数文件（从主库修改得来）
orcl.__db_cache_size=2063597568
orcl.__java_pool_size=16777216
orcl.__large_pool_size=33554432
orcl.__oracle_base='C:\app\Administrator'#ORACLE_BASE set from environment
orcl.__pga_aggregate_target=872415232
orcl.__sga_target=2583691264
orcl.__shared_io_pool_size=0
orcl.__shared_pool_size=436207616
orcl.__streams_pool_size=0
*.audit_file_dest='C:\app\Administrator\admin\orcl\adump'
*.audit_trail='db'
*.compatible='11.2.0.4.0'
*.control_files='D:\oradata\orcldg\control01.ctl','D:\oradata\orcldg\control02.ctl'
*.db_block_size=8192
*.db_domain=''
*.db_file_name_convert='d:\oradata\orcl','d:\oradata\orcldg'
*.db_name='orcl'
*.db_unique_name='orcldg'
*.db_recovery_file_dest='E:\flash_recovery_area'
*.db_recovery_file_dest_size=43851448320
*.diagnostic_dest='C:\app\Administrator'
*.dispatchers='(PROTOCOL=TCP) (SERVICE=orclXDB)'
*.fal_server='ORCL'
*.log_archive_config='DG_CONFIG=(orcl,orcldg)'
*.log_archive_dest_1='LOCATION=USE_DB_RECOVERY_FILE_DEST VALID_FOR=(ALL_LOGFILES,ALL_ROLES) DB_UNIQUE_NAME=orcldg'
*.log_archive_dest_2='SERVICE=orcl ASYNC VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=orcl'
*.log_archive_format='ARC%S_%R.%T'
*.log_file_name_convert='d:\oradata\orcl','d:\oradata\orcldg'
*.nls_language='SIMPLIFIED CHINESE'
*.nls_territory='CHINA'
*.open_cursors=300
*.pga_aggregate_target=858783744
*.processes=1500
*.remote_login_passwordfile='EXCLUSIVE'
*.sessions=1655
*.sga_target=2576351232
*.standby_file_management='AUTO'
*.undo_tablespace='UNDOTBS1'
3、创建备库实例
C:\Users\Administrator>oradim -new -sid orcl -startmode m
4、监听listener.ora、tnsnames.ora文件
listener.ora
LISTENER =
(DESCRIPTION_LIST =
(DESCRIPTION =
(ADDRESS = (PROTOCOL = TCP)(HOST = 192.168.10.200)(PORT = 1521))
)
)
SID_LIST_LISTENER =
(SID_LIST =
(SID_DESC =
(SID_NAME = CLRExtProc)
(ORACLE_HOME = C:\app\Administrator\product\11.2.0\dbhome_1)
(PROGRAM = extproc)
(ENVS = "EXTPROC_DLLS=ONLY:C:\app\Administrator\product\11.2.0\dbhome_1\bin\oraclr11.dll")
)
(SID_DESC =
(GLOBAL_DBNAME = orcl)
(SID_NAME = orcl)
)
)
ADR_BASE_LISTENER = C:\app\Administrator
tnsnames.ora
ORACLR_CONNECTION_DATA =
(DESCRIPTION =
(ADDRESS_LIST =
(ADDRESS = (PROTOCOL = IPC)(KEY = EXTPROC1521))
)
(CONNECT_DATA =
(SID = CLRExtProc)
(PRESENTATION = RO)
)
)
orcl =
(DESCRIPTION =
(ADDRESS = (PROTOCOL = TCP)(HOST = 192.168.10.100)(PORT = 1521))
(CONNECT_DATA =
(SERVER = DEDICATED)
(SERVICE_NAME = orcl)
)
)
orcldg =
(DESCRIPTION =
(ADDRESS = (PROTOCOL = TCP)(HOST = 192.168.10.200)(PORT = 1521))
(CONNECT_DATA =
(SERVER = DEDICATED)
(SERVICE_NAME = orcl)
)
)
4、密码文件（复制主库密码文件）
windows密码文件位置：%ORACLE_HOME%/database/
linux密码文件位置：$ORACLE_HOME/dbs/
5、创建参数文件并启动到mount状态
6、备份主库及备库控制文件到备库
（1）为备库创建控制文件standby.ctl
SQL> alter database create standby controlfile as 'd:\standby.ctl';
（2）备份现有主库
backup.bat
@echo off
set timestamp=%date:~0,4%%date:~5,2%%date:~8,2%
mkdir d:\backup\%timestamp%
rman target / nocatalog CMDFILE 'd:\scripts\full_backup.sql' log 'd:\scripts\log\rman_backup_%timestamp%.log' append
full_backup.sql
run{
CONFIGURE DEFAULT DEVICE TYPE TO DISK; # default
CONFIGURE CONTROLFILE AUTOBACKUP ON;#
allocate channel ch00 device type disk;
allocate channel ch01 device type disk;
allocate channel ch02 device type disk;
allocate channel ch03 device type disk;
crosscheck backup;
crosscheck archivelog all;
delete noprompt expired archivelog all;
delete noprompt expired backup;
report obsolete;
delete noprompt obsolete;
backup as compressed backupset database format 'D:\backup\%T\full_%T_%u_%p.%d' tag='FULLDB-%T'
plus archivelog format 'D:\backup\%T\arch_%T_%u_%p.%d' delete all input;
sql 'alter system archive log current';
backup current controlfile format 'D:\backup\%T\control_%T_%u_%p.%d.bak';
backup spfile format 'D:\backup\%T\spfile_%T_%u_%p.%d.bak';
crosscheck backup;
crosscheck archivelog all;
delete noprompt expired archivelog all;
delete noprompt expired backup;
report obsolete;
delete noprompt obsolete;
release channel ch00;
release channel ch01;
release channel ch02;
release channel ch03;
}
（3）拷贝备库控制文件及备份文件到备库
7、恢复备库
（1）恢复控制文件
（2）恢复数据文件
RMAN> catalog start with 'E:\20260701\';
RMAN> restore database;
（3）备库应用日志同步
alter database recover managed standby database using current logfile disconnect from session; (同步主库日志)备库mount状态第一次同步主备数据；
alter database recover managed standby database cancel;(取消同步)
Alter database open；（启动备数据库到open read only模式）
alter database recover managed standby database using current logfile disconnect from session;（再次应用日志同步）
（4）检查主备数据状态
主库：SQL> select error,status from v$archive_dest where dest_id=2;
备注：主库日志传输到备库通道2有效；
备库：SQL> select process,sequence#,status from v$managed_standby;
备注：检查备库应用日志状态，显示应用日志正常。
（5）主库录入测试数据，检查备库情况
主库：
备库：
以上步骤不影响主库业务正常运行，接下来的步骤影响业务运行，需要申请停机时间。（30分钟）
四、主备切换
1、检查主库状态（原历史库）
备注：session active表示有终端还在连接主库，此时需要断开连接，到to standby状态才能切换。
2、检查备库状态
3、切换主库到备库（主库业务中断）
SQL> alter database commit to switchover to standby with session shutdown ;
4、备库切换到主库
SQL> alter database commit to switchover to primary;
新备库应用日志：
SQL> alter database recover managed standby database using current logfile disconnect;
5、检查新的主备状态
（1）主库录入数据，测试备库数据同步情况
主库：
备库：
五、带入双机集群
1、停用现有主库，手动脱盘d盘和E盘
2、通过集群软件带入双机
3、检查双机状态
4、检查双机tnsnames.ora和监听文件（确保客户端可以正常连接）
5、来回切换检查数据库是否正常
六、新双机集群搭建完成，去除adg参数
1、参数（停机删除）
修改后参数文件：
orcl.__db_cache_size=1493172224
orcl.__java_pool_size=16777216
orcl.__large_pool_size=603979776
orcl.__oracle_base='c:\app\Administrator'#ORACLE_BASE set from environment
orcl.__pga_aggregate_target=872415232
orcl.__sga_target=2583691264
orcl.__shared_io_pool_size=0
orcl.__shared_pool_size=436207616
orcl.__streams_pool_size=0
*.audit_file_dest='C:\app\Administrator\admin\orcl\adump'
*.audit_trail='db'
*.compatible='11.2.0.4.0'
*.control_files='D:\ORADATA\ORCLDG\CONTROL01.CTL','D:\ORADATA\ORCLDG\CONTROL02.CTL'#Restore Controlfile
*.db_block_size=8192
*.db_domain=''
*.db_file_name_convert='d:\oradata\orcl','d:\oradata\orcldg'
*.db_name='orcl'
*.db_recovery_file_dest='E:\flash_recovery_area'
*.db_recovery_file_dest_size=43851448320
*.db_unique_name='orcldg'
*.diagnostic_dest='C:\app\Administrator'
*.dispatchers='(PROTOCOL=TCP) (SERVICE=orclXDB)'
*.log_archive_format='ARC%S_%R.%T'
*.nls_language='SIMPLIFIED CHINESE'
*.nls_territory='CHINA'
*.open_cursors=300
*.pga_aggregate_target=858783744
*.processes=1500
*.remote_login_passwordfile='EXCLUSIVE'
*.sessions=1655
*.sga_target=2576351232
*.undo_tablespace='UNDOTBS1'
2、启动主库
备注：切换无日志告警，双机切换完成，adg切换到集群主库完成。
总结：实验中遇到有3个问题如下；
1、winows数据库安装软件完成后，手动创建实例，否则无法登录sqlplus / as sysdba;
2、备库密码文件名称问题导致主备应用日志无法同步，要求和实例名称相同即可；
3、adg同步库切换到单主库模式需要停机，对参数文件中有adg参数的一一删除，其他standbylog等信息可以不删，不影响业务使用。
2026-07-01
于合肥