配置Oracle 11g datagurad redo日志cascade传输

目标：实现dataguard redo传输cascade，即dg1的redo日志先传输到dg2，然后再从dg2传输到dg3,从而减少dg1主库的网络带宽压力！通常使用于dg1和dg2在同一局域网内，dg2和dg3通过广域网相连的场合！

一：环境介绍
主库
IP地址:192.168.1.61/24
操作系统版本：rhel5.4 64bit
数据库版本：11.2.0.3 64bit
数据库sid名：dg
数据库名：dg
数据库db_unique_name:dg1

备库1  物理备库
IP地址:192.168.1.62/24
操作系统版本：rhel5.4 64bit
数据库版本：11.2.0.3 64bit
数据库sid名：dg
数据库名：dg
数据库db_unique_name:dg2

备库2  物理备库
IP地址:192.168.1.62/24
操作系统版本：rhel5.4 64bit
数据库版本：11.2.0.3 64bit
数据库sid名：dg
数据库名：dg
数据库db_unique_name:dg3

二：统一修改三台数据库服务器的listener.ora和tnsnames.ora文件如下,listener.ora文件里面的GLOBAL_DBNAME和HOST参数做相应的修改

[oracle@dg1 ~]$ cat $TNS_ADMIN/listener.ora SID_LIST_LISTENER =   (SID_LIST =     (SID_DESC =       (GLOBAL_DBNAME = dg1.yang.com)       (ORACLE_HOME = /u01/app/oracle/product/11.2.0/db1)       (SID_NAME = dg)     )   ) LISTENER =   (DESCRIPTION_LIST =     (DESCRIPTION =       (ADDRESS_LIST =         (ADDRESS = (PROTOCOL = TCP)(HOST = dg1.yang.com)(PORT = 1521))       )     )   ) [oracle@dg1 ~]$ grep -v '^#' $TNS_ADMIN/tnsnames.ora dg1 =   (DESCRIPTION =     (ADDRESS = (PROTOCOL = TCP)(HOST = 192.168.1.61)(PORT = 1521))     (CONNECT_DATA =       (SERVER = DEDICATED)       (SERVICE_NAME = dg1.yang.com)     )   )   dg2 =   (DESCRIPTION =     (ADDRESS = (PROTOCOL = TCP)(HOST = 192.168.1.62)(PORT = 1521))     (CONNECT_DATA =       (SERVER = DEDICATED)       (SERVICE_NAME = dg2.yang.com)     )   ) dg3 =   (DESCRIPTION =     (ADDRESS = (PROTOCOL = TCP)(HOST = 192.168.1.63)(PORT = 1521))     (CONNECT_DATA =       (SERVER = DEDICATED)       (SERVICE_NAME = dg3.yang.com)     )   )

三：dg1和dg2数据库服务器构建active dataguard
3.1修改主库初始化参数,添加standby日志组

[oracle@dg1 ~]$ sqlplus /nolog
SQL*Plus: Release 11.2.0.3.0 Production on Sat Apr 7 14:11:18 2012
Copyright (c) 1982, 2011, Oracle.  All rights reserved.

SQL> conn /as sysdba
Connected.

SQL> !mkdir /u01/app/oracle/archivelog

SQL> alter system set log_archive_dest_1='LOCATION=/u01/app/oracle/archivelog'
valid_for=(all_logfiles,primary_role) db_unique_name=dg1;
System altered.

SQL> archive log list;
Database log mode              Archive Mode
Automatic archival             Enabled
Archive destination            /u01/app/oracle/archivelog
Oldest online log sequence     3
Next log sequence to archive   5
Current log sequence           5

SQL> select force_logging,db_unique_name from v$database;

FOR DB_UNIQUE_NAME
--- ------------------------------
NO  dg

SQL> alter database force logging;
Database altered.

SQL> alter system set db_unique_name='dg1' scope=spfile;
System altered.

SQL> select force_logging,db_unique_name from v$database;

FOR DB_UNIQUE_NAME
--- ------------------------------
YES dg

SQL>alter system set log_archive_config='DG_CONFIG=(dg1,dg2,dg3)';
System altered.

SQL>alter system set log_archive_dest_2='SERVICE=dg2 valid_for=(online_logfile,primary_role) db_unique_name=dg2' scope=spfile;
System altered.

SQL>alter system set log_archive_dest_3='SERVICE=dg3  valid_for=(standby_logfiles,standby_role) db_unique_name=dg3' scope=spfile;
System altered.


SQL> show parameter remote_login;

NAME                                 TYPE        VALUE
------------------------------------ ----------- ------------------------------
remote_login_passwordfile            string      EXCLUSIVE

SQL> show parameter log_archive_format;

NAME                                 TYPE        VALUE
------------------------------------ ----------- ------------------------------
log_archive_format                   string      %t_%s_%r.dbf

SQL> alter database add standby logfile group 4  '/u01/app/oracle/oradata/dg/standby01.log' size 50M;
Database altered.

SQL> alter database add standby logfile group 5  '/u01/app/oracle/oradata/dg/standby02.log' size 50M;
Database altered.

SQL> alter database add standby logfile group 6  '/u01/app/oracle/oradata/dg/standby03.log' size 50M;
Database altered.

SQL> alter database add standby logfile group 7  '/u01/app/oracle/oradata/dg/standby04.log' size 50M;
Database altered.

SQL> select member,group# from v$logfile order by 2;

MEMBER                                       GROUP#
---------------------------------------- ----------
/u01/app/oracle/oradata/dg/redo01.log             1
/u01/app/oracle/oradata/dg/redo02.log             2
/u01/app/oracle/oradata/dg/redo03.log             3
/u01/app/oracle/oradata/dg/standby01.log          4
/u01/app/oracle/oradata/dg/standby02.log          5
/u01/app/oracle/oradata/dg/standby03.log          6
/u01/app/oracle/oradata/dg/standby04.log          7
7 rows selected.

一：环境介绍3. 3.2 由spfile生成pfile,重启主库，使先前修改的参数生效!修改参数文件，scp至dg2

SQL> create pfile='/home/oracle/initdg.ora' from spfile;
File created.

SQL> shutdown immediate
Database closed.
Database dismounted.
ORACLE instance shut down.

SQL> startup
ORACLE instance started.

Total System Global Area  417546240 bytes
Fixed Size                  2228944 bytes
Variable Size             281021744 bytes
Database Buffers          125829120 bytes
Redo Buffers                8466432 bytes
Database mounted.
Database opened.

SQL> show parameter name;

NAME                                 TYPE        VALUE
------------------------------------ ----------- ------------------------------
db_file_name_convert                 string
db_name                              string      dg
db_unique_name                       string      dg1
global_names                         boolean     FALSE
instance_name                        string      dg
lock_name_space                      string
log_file_name_convert                string
processor_group_name                 string
service_names                        string      dg1.yang.com

[oracle@dg1 ~]$ cat initdg.ora
dg.__db_cache_size=125829120
dg.__java_pool_size=4194304
dg.__large_pool_size=4194304
dg.__oracle_base='/u01/app/oracle'#ORACLE_BASE set from environment
dg.__pga_aggregate_target=167772160
dg.__sga_target=251658240
dg.__shared_io_pool_size=0
dg.__shared_pool_size=104857600
dg.__streams_pool_size=0
*.audit_file_dest='/u01/app/oracle/admin/dg/adump'
*.audit_trail='db'
*.compatible='11.2.0.0.0'
*.control_files='/u01/app/oracle/oradata/dg/control01.ctl','/u01/app/oracle/fast_recovery_area/dg/control02.ctl'
*.db_block_size=8192
*.db_domain='yang.com'
*.db_name='dg'
*.db_recovery_file_dest='/u01/app/oracle/fast_recovery_area'
*.db_recovery_file_dest_size=4322230272
*.db_unique_name='dg2'
*.diagnostic_dest='/u01/app/oracle'
*.dispatchers='(PROTOCOL=TCP) (SERVICE=dgXDB)'
*.log_archive_config='DG_CONFIG=(dg1,dg2,dg3)'
*.log_archive_dest_1='LOCATION=/u01/app/oracle/archivelog valid_for=(all_logfiles,primary_role) db_unique_name=dg2'
*.log_archive_dest_2='SERVICE=dg1  valid_for=(online_logfiles,primary_role) db_unique_name=dg1'
*.log_archive_dest_3='SERVICE=dg3  valid_for=(standby_logfiles,standby_role) db_unique_name=dg3'
*.log_archive_format='%t_%s_%r.dbf'
*.memory_target=419430400
*.open_cursors=300
*.processes=150
*.remote_login_passwordfile='EXCLUSIVE'
*.undo_tablespace='UNDOTBS1

[oracle@dg1 ~]$ scp initdg.ora 192.168.1.62:/home/oracle/

3.3 在dg2上创建密码文件，由pfile生成spfile文件，启动数据库实例到nomount状态，通过rman进行duplicate操作，在11g中，duplicate数据库可以from active database，意味着不需要先在主库上进行rman备份，然后再将备份文件复制到备库上，在10g版本上需要这样操作，操作完成后将备库1置于active dataguard状态，即实时应用日志并只读的状态，adg也是11g dataguard非常大的一个飞跃(dg2只安装数据库软件，未建库)

[oracle@dg2 ~]$ orapwd file=$ORACLE_HOME/dbs/orapwdg password=123456 entries=5
[oracle@dg2 ~]$ sqlplus /nolog
SQL*Plus: Release 11.2.0.3.0 Production on Sat Apr 7 15:40:33 2012
Copyright (c) 1982, 2011, Oracle.  All rights reserved.

SQL> conn /as sysdba
Connected to an idle instance.
SQL> create spfile from pfile='/home/oracle/initdg.ora';
File created.

SQL> startup nomount
ORACLE instance started.

Total System Global Area  417546240 bytes
Fixed Size                  2228944 bytes
Variable Size             281021744 bytes
Database Buffers          125829120 bytes
Redo Buffers                8466432 bytes

[oracle@dg2 ~]$ rman target  sys/123456@dg1 auxiliary sys/123456@dg2
Recovery Manager: Release 11.2.0.3.0 - Production on Sat Apr 7 16:22:57 2012
Copyright (c) 1982, 2011, Oracle and/or its affiliates.  All rights reserved.
connected to target database: DG (DBID=1693312161)
connected to auxiliary database: DG (not mounted)

RMAN> list backup of database summary; //这个时候主库没有任何的物理备份
using target database control file instead of recovery catalog
specification does not match any backup in the repository

RMAN> duplicate target database for standby nofilenamecheck from active database;

————————输出省略——————

[oracle@dg2 ~]$ sqlplus /nolog
SQL*Plus: Release 11.2.0.3.0 Production on Sat Apr 7 16:30:11 2012
Copyright (c) 1982, 2011, Oracle.  All rights reserved.

SQL> conn /as sysdba
Connected.
SQL> select database_role,open_mode,name from v$database;

DATABASE_ROLE    OPEN_MODE            NAME
---------------- -------------------- ---------
PHYSICAL STANDBY MOUNTED              DG

SQL> archive log list;
Database log mode              Archive Mode
Automatic archival             Enabled
Archive destination            /u01/app/oracle/archivelog
Oldest online log sequence     8
Next log sequence to archive   0
Current log sequence           10

SQL> alter database open read only;
Database altered.

SQL> alter database recover managed standby database using current logfile disconnect from session;
Database altered.

SQL> select database_role,open_mode,name from v$database;

DATABASE_ROLE    OPEN_MODE            NAME
---------------- -------------------- ---------
PHYSICAL STANDBY READ ONLY WITH APPLY DG

3.4测试主库上切换日志

SQL> alter system switch logfile;
System altered.

SQL> select message from v$dataguard_status;
MESSAGE
--------------------------------------------------------------------------------
LNS: Completed archiving log 3 thread 1 sequence 9
ARC0: Beginning to archive thread 1 sequence 9 (1049579-1050098)
ARC0: Completed archiving thread 1 sequence 9 (1049579-1050098)
LNS: Standby redo logfile selected for thread 1 sequence 10 for destination LOG_
ARCHIVE_DEST_2

LNS: Beginning to archive log 1 thread 1 sequence 10
LNS: Completed archiving log 1 thread 1 sequence 10
ARC2: Beginning to archive thread 1 sequence 10 (1050098-1050271)
ARC2: Completed archiving thread 1 sequence 10 (1050098-1050271)
LNS: Beginning to archive log 2 thread 1 sequence 11

备库上查看日志应用情况
SQL> archive log list;
Database log mode              Archive Mode
Automatic archival             Enabled
Archive destination            /u01/app/oracle/archivelog
Oldest online log sequence     9
Next log sequence to archive   0
Current log sequence           11

3.5 实际操作应用情况，发现需要主库上切换日志后，数据才会同步，也就是standby日志组未使用

SQL> create table t1 as select * from dba_objects;
Table created.

SQL> alter system switch logfile;
System altered.

SQL> conn sys/123456@dg2 as sysdba
Connected.
SQL> select count(*) from t1;

  COUNT(*)
----------
     74510

四：配置redo传输的cascade
4.1 将修改过的初始化文件传输到dg3上(dg3服务器上同样只安装数据库软件，不建库)

[oracle@dg2 ~]$ cat initdg.ora dg.__db_cache_size=125829120 dg.__java_pool_size=4194304 dg.__large_pool_size=4194304 dg.__oracle_base='/u01/app/oracle'#ORACLE_BASE set from environment dg.__pga_aggregate_target=167772160 dg.__sga_target=251658240 dg.__shared_io_pool_size=0 dg.__shared_pool_size=104857600 dg.__streams_pool_size=0 *.audit_file_dest='/u01/app/oracle/admin/dg/adump' *.audit_trail='db' *.compatible='11.2.0.0.0' *.control_files='/u01/app/oracle/oradata/dg/control01.ctl','/u01/app/oracle/fast_recovery_area/dg/control02.ctl' *.db_block_size=8192 *.db_domain='yang.com' *.db_name='dg' *.db_recovery_file_dest='/u01/app/oracle/fast_recovery_area' *.db_recovery_file_dest_size=4322230272 *.db_unique_name='dg3' *.diagnostic_dest='/u01/app/oracle' *.dispatchers='(PROTOCOL=TCP) (SERVICE=dgXDB)' *.log_archive_config='DG_CONFIG=(dg1,dg2,dg3)' *.log_archive_dest_1='LOCATION=/u01/app/oracle/archivelog valid_for=(all_logfiles,primary_role) db_unique_name=dg3' *.standby_archive_dest=/u01/app/oracle/standbylog *.log_archive_format='%t_%s_%r.dbf' *.memory_target=419430400 *.open_cursors=300 *.processes=150 *.remote_login_passwordfile='EXCLUSIVE' *.undo_tablespace='UNDOTBS1'

4.2 通过pfile生成spfile,将数据库启动到nomount状态，然后进行duplicate

[oracle@dg3 ~]$ sqlplus /nolog
SQL*Plus: Release 11.2.0.3.0 Production on Sat Apr 14 18:01:59 2012
Copyright (c) 1982, 2011, Oracle.  All rights reserved.

SQL> conn /as sysdba
Connected to an idle instance.
SQL> create spfile from pfile='/home/oracle/initdg.ora';
File created.

SQL> startup nomount
ORACLE instance started.

Total System Global Area  417546240 bytes
Fixed Size                  2228944 bytes
Variable Size             281021744 bytes
Database Buffers          125829120 bytes
Redo Buffers                8466432 bytes

[oracle@dg3 ~]$ rman target  sys/123456@dg1 auxiliary sys/123456@dg3
Recovery Manager: Release 11.2.0.3.0 - Production on Sat Apr 14 18:02:59 2012
Copyright (c) 1982, 2011, Oracle and/or its affiliates.  All rights reserved.

connected to target database: DG (DBID=1693928392)
connected to auxiliary database: DG (not mounted)

RMAN> duplicate target database for standby nofilenamecheck from active database;

————————输出省略——————

 4.3 将备库dg3置于只读且应用日志模式下

[oracle@dg3 ~]$ sqlplus /nolog
SQL*Plus: Release 11.2.0.3.0 Production on Sat Apr 14 18:49:12 2012
Copyright (c) 1982, 2011, Oracle.  All rights reserved.

SQL> conn /as sysdba
Connected.
SQL> alter database open read only;
Database altered.

SQL> alter database recover managed standby database using current logfile disconnect from session;
Database altered.

SQL> select open_mode,database_role,name from v$database;

OPEN_MODE            DATABASE_ROLE    NAME
-------------------- ---------------- ---------
READ ONLY WITH APPLY PHYSICAL STANDBY DG

五：测试

SQL> conn /as sysdba
Connected.
SQL> select open_mode,database_role,name from v$database;

OPEN_MODE            DATABASE_ROLE    NAME
-------------------- ---------------- ---------
READ WRITE           PRIMARY          DG

SQL> truncate table t1;
Table truncated.

SQL> conn sys/123456@dg2 as sysdba
Connected.
SQL> select count(*) from t1;

  COUNT(*)
----------
     74510

SQL> conn sys/123456@dg3 as sysdba
Connected.
SQL> select count(*) from t1;

  COUNT(*)
----------
     74510

SQL> conn /as sysdba
Connected.
SQL> alter system switch logfile; //主库上未切换日志前，数据不能同步
System altered.

SQL> conn sys/123456@dg2 as sysdba
Connected.
SQL> select count(*) from t1;

  COUNT(*)
----------
         0

SQL> conn sys/123456@dg3 as sysdba  //在dg3上可以看到明显的延迟情况
Connected.
SQL> select count(*) from t1;

  COUNT(*)
----------
     74510

SQL> select count(*) from t1;

  COUNT(*)
----------
         0

SQL> select message from v$dataguard_status; //查看备库dg3的dataguard状态
MESSAGE
--------------------------------------------------------------------------------
Media Recovery Log /u01/app/oracle/standbylog/1_13_780597393.dbf
Media Recovery Waiting for thread 1 sequence 14
FAL[client]: Error fetching gap sequence, no FAL server specified
RFS[1]: Assigned to RFS process 2701
Media Recovery Log /u01/app/oracle/archivelog/1_14_780597393.dbf
Media Recovery Waiting for thread 1 sequence 15
FAL[client]: Error fetching gap sequence, no FAL server specified
18 rows selected.

SQL> alter system set fal_server='dg2';
System altered.

SQL> alter system set fal_client='dg3';
System altered.

总结：
1：在测试上，可以将主库的log_archive_dest_state_3参数设置为defer，然后切换日志，观察dg2，dg3的日志传输情况
2：在本文档中没有设置standby_file_management参数，因为不涉及新建数据文件的操作
3：在dg1和dg2上均未设置fal_server和fal_client参数，因为此处不考虑dg的角色切换问题，纯粹测试redo日志的cascade传输
4：本文档未能成功实现adg，猜想和几个归档日志路径配置的参数相关，后续将继续测试！

本文出自 “斩月” 博客，谢绝转载！