DB / Oracle / 安装部署

Oracle Data Guard 部署

2026.09.06

Oracle Data Guard 部署

适用版本：Oracle Database 19c 文档版本：v1.0 / 2026-07

1. 概述

Data Guard 提供灾备[1]：

类型：

  • Physical Standby
  • Logical Standby
  • Snapshot Standby
  • Far Sync

详细见：Oracle Data Guard 容灾。

2. 主库准备

2.1 归档模式

SHUTDOWN IMMEDIATE;
STARTUP MOUNT;
ALTER DATABASE ARCHIVELOG;
ALTER DATABASE OPEN;

ARCHIVE LOG LIST;

2.2 Force Logging

ALTER DATABASE FORCE LOGGING;

2.3 Standby Redo Log

-- 组数 = (Redo 组数 + 1)
ALTER DATABASE ADD STANDBY LOGFILE GROUP 11 
  ('/u01/oradata/stdby_redo11.log') SIZE 200M;
ALTER DATABASE ADD STANDBY LOGFILE GROUP 12 
  ('/u01/oradata/stdby_redo12.log') SIZE 200M;
ALTER DATABASE ADD STANDBY LOGFILE GROUP 13 
  ('/u01/oradata/stdby_redo13.log') SIZE 200M;
ALTER DATABASE ADD STANDBY LOGFILE GROUP 14 
  ('/u01/oradata/stdby_redo14.log') SIZE 200M;

2.4 参数

ALTER SYSTEM SET log_archive_config = 'dg_config=(orcl,orcldg)';
ALTER SYSTEM SET log_archive_dest_1 = 'location=use_db_recovery_file_dest valid_for=(all_logfiles,all_roles) db_unique_name=orcl';
ALTER SYSTEM SET log_archive_dest_2 = 'service=orcldg async valid_for=(online_logfiles,primary_role) db_unique_name=orcldg';
ALTER SYSTEM SET log_archive_dest_state_1 = enable;
ALTER SYSTEM SET log_archive_dest_state_2 = enable;
ALTER SYSTEM SET fal_server = 'orcldg';
ALTER SYSTEM SET fal_client = 'orcl';
ALTER SYSTEM SET standby_file_management = auto;
ALTER SYSTEM SET db_file_name_convert = '/u01/oradata/orcldg/', '/u01/oradata/orcl/';
ALTER SYSTEM SET log_file_name_convert = '/u01/oradata/orcldg/', '/u01/oradata/orcl/';

3. 网络配置

3.1 tnsnames.ora

ORCL = 
  (DESCRIPTION = 
    (ADDRESS = (PROTOCOL = tcp)(HOST = primary)(PORT = 1521))
    (CONNECT_DATA = (SERVICE_NAME = orcl))
  )

ORCLDG = 
  (DESCRIPTION = 
    (ADDRESS = (PROTOCOL = tcp)(HOST = standby)(PORT = 1521))
    (CONNECT_DATA = (SERVICE_NAME = orcldg))
  )

3.2 静态注册（备库）

# listener.ora
SID_LIST_LISTENER = (SID_LIST = (SID_DESC = (GLOBAL_DBNAME = orcldg)(ORACLE_HOME = ...)(SID_NAME = orcldg)))

详细见：Oracle 网络配置与监听器。

4. 备库准备

4.1 密码文件

# 主库
scp $ORACLE_HOME/dbs/orapworcl standby:$ORACLE_HOME/dbs/orapworcldg

4.2 参数文件

-- 主库
CREATE PFILE='/tmp/initorcldg.ora' FROM SPFILE;
scp /tmp/initorcldg.ora standby:/tmp/

-- 备库调整
db_unique_name='orcldg'
log_archive_dest_2='service=orcl async valid_for=(online_logfiles,primary_role) db_unique_name=orcl'
fal_server='orcl'
fal_client='orcldg'

4.3 创建目录

mkdir -p ******
mkdir -p ******
mkdir -p ******

5. RMAN Duplicate

5.1 启动备库到 NOMOUNT

sqlplus / as sysdba
SQL> STARTUP NOMOUNT PFILE='?/dbs/initorcldg.ora';

5.2 RMAN Duplicate

rman target sys/pwd@orcl auxiliary sys/pwd@orcldg

RMAN> DUPLICATE TARGET DATABASE 
  FOR STANDBY 
  FROM ACTIVE DATABASE 
  DORECOVER 
  NOFILENAMECHECK;

6. 启动应用

6.1 创建 SPFILE

CREATE SPFILE FROM PFILE;
SHUTDOWN IMMEDIATE;
STARTUP MOUNT;

6.2 启动 MRP

ALTER DATABASE RECOVER MANAGED STANDBY DATABASE 
  DISCONNECT FROM SESSION USING CURRENT LOGFILE;

6.3 开启 Active Data Guard

ALTER DATABASE RECOVER MANAGED STANDBY DATABASE CANCEL;
ALTER DATABASE OPEN READ ONLY;
ALTER DATABASE RECOVER MANAGED STANDBY DATABASE 
  DISCONNECT FROM SESSION USING CURRENT LOGFILE;

详细见：Oracle Data Guard 性能优化。

7. 验证

7.1 主库

SELECT dest_id, status, error FROM v$archive_dest WHERE dest_id = 2;
SELECT thread#, sequence#, archived, applied FROM v$archived_log ORDER BY sequence# DESC;

7.2 备库

SELECT process, status, sequence# FROM v$managed_standby;
SELECT database_role, open_mode, protection_mode FROM v$database;
SELECT name, value, unit FROM v$dataguard_stats;

7.3 日志

# 主库
ALTER SYSTEM SWITCH LOGFILE;
# 备库
SELECT sequence#, applied FROM v$archived_log ORDER BY sequence# DESC;

8. Switchover

8.1 主库验证

SELECT switchover_status FROM v$database;
-- TO STANDBY

8.2 主库切换

ALTER DATABASE COMMIT TO SWITCHOVER TO STANDBY WITH SESSION SHUTDOWN;
SHUTDOWN IMMEDIATE;
STARTUP NOMOUNT;
ALTER DATABASE MOUNT STANDBY DATABASE;
ALTER DATABASE RECOVER MANAGED STANDBY DATABASE DISCONNECT;

8.3 备库切换

ALTER DATABASE COMMIT TO SWITCHOVER TO PRIMARY;
SHUTDOWN IMMEDIATE;
STARTUP;

9. Failover

9.1 备库停止应用

ALTER DATABASE RECOVER MANAGED STANDBY DATABASE FINISH;

9.2 切换

ALTER DATABASE ACTIVATE STANDBY DATABASE;
SHUTDOWN IMMEDIATE;
STARTUP;

9.3 重建原主库

- Flashback Database
- RMAN 重建

10. 监控

10.1 延迟

SELECT name, value, unit, time_computed 
FROM v$dataguard_stats;

10.2 GAP

SELECT * FROM v$archive_gap;

10.3 进程

SELECT process, status, sequence#, block#, blocks 
FROM v$managed_standby;

11. 常见坑与排错

11.1 GAP

-- 1. 查找
SELECT * FROM v$archive_gap;

-- 2. 注册
ALTER DATABASE REGISTER LOGFILE '...';

11.2 备库落后

# 1. 网络带宽
# 2. 备库 I/O
# 3. 并行应用
ALTER DATABASE RECOVER MANAGED STANDBY DATABASE PARALLEL 8;

11.3 ORA-12154

# 1. TNS 配置
# 2. 监听
# 3. 服务名

12. 最佳实践

  1. 归档模式：基础
  2. Force Logging：完整
  3. Standby Redo Log：实时
  4. ASYNC：性能
  5. Active Data Guard：备库利用
  6. TAF：业务切换
  7. 监控延迟：及时
  8. GAP 处理：完整
  9. 定期演练：验证
  10. 文档化：流程

13. 参考资料

[1] Oracle Data Guard Concepts and Administration 19c https://docs.oracle.com/en/database/oracle/oracle-database/19/sbydb/