【DG实验】搭建physical standby——rman

使用RMAN 复制搭建物理Data Guard

生产库搭建物理DG一般使用rman而不是copy。Rman若不需要修改db_name, db_unique_name这两个参数，就可以完全不停业务进行搭建物理DG，而copy需要停业务，主库一致处于mount状态！

 

1.DG环境说明

数据库版本：11.2.0.1.0

主库：192.168.88.3 sharon

备库：192.168.88.2 sharon

使用静态IP，要确保IP是固定的，不要用DHCP来分配！

 

Database      DB_UNIQUE_NAME   Oracle Net Service Name

Primary                      sharon                        sharon_pd

Physical standby         sharon                        sharon_st

 

2.主库设置为force logging 模式

SQL> alter database force logging;

SQL> select FORCE_LOGGING from v$database;

FORCE_LOGGING

--------------------------  

YES

 

3. 主库修改为归档模式

SQL> archive log list; 
SQL> shutdown immediate 
SQL> startup mount 
SQL> alter database archivelog; 
SQL> alter system set log_archive_dest_1='location=/u01/arch/' scope=spfile;
SQL> archive log list; 

 

4. 在主备库分别创建Listener，并启动

用netca或netmgr配

启动监听

[oracle@sharon admin]$lsnrctl status|start|stop

此步若不清楚，可参考用copy file方式配置物理DG的部分。

http://blog.csdn.net/sharqueen/article/details/8853118

 

5. 在主备库添加Oracle Net Service（tnsnames.ora）

注意：

sharon_pd,sharon_st——去连接的，不代表实例名

GLOBAL_DBNAME = sharon这个名可以随便改

tsnping下，是否通

[oracle@sharon admin]$tnsping sharon_pd

[oracle@sharon admin]$tnsping sharon_st

 

6. 在主库创建pfile 文件并修改pfile 内容

SQL> create pfile from spfile;

在pfile添加如下内容：
#add for primary dg
*.db_unique_name='sharon_pd' 
*.log_archive_config='dg_config=(sharon_pd,sharon_st)'
*.log_archive_dest_1='location=/u01/archlog valid_for=(all_logfiles,all_roles) db_unique_name=sharon_pd'
*.log_archive_dest_2='service=sharon_st reopen=120 lgwr async valid_for=(online_logfiles,primary_role) db_unique_name=sharon_st'
*.log_archive_dest_state_1=enable 
*.log_archive_dest_state_2=enable 
*.standby_file_management='auto'
*.fal_server='sharon_st'

*.fal_client=’sharon_pd’   –10g 要写！11g不需要写！

注意： 
（1）在oracle 11g中，废弃了fal_client 参数，也就是说不用配置。
（2）log_archive_dest_n 这个参数中的Oracle Net Service名首位是一样的，前面写哪个，最后的db_unique_name就写哪个，不要搞错了。
（3）如果主备库的路径不同，修改在主库的参数文件里添加如下2个参数：
*.log_file_name_convert='/u02/oradata/orcl/','/u03/oradata/orcl/'
*.db_file_name_convert='/u02/oradata/orcl/','/u03/oradata/orcl/'

 

--用修改过的pfile重新生成spfile：

SQL>startup pfile=’$ORACLE_HOME/dbs/initsharon.ora’;
SQL> create spfile from pfile;

 

7. 在备库创建相关的目录

参考pfile，不要漏了

[oracle@sharon sharon]$ mkdir –p /u01/app/oracle/oradata/sharon

[oracle@sharon sharon]$ mkdir –p /u01/app/oracle/fast_recovery_area/sharon

[oracle@sharon sharon]$ mkdir –p /u01/app/oracle/admin/sharon/adump

[oracle@sharon sharon]$ mkdir –p /u01/app/oracle/admin/sharon/dpdump

 

8. 将主库的参数文件copy到备库并修改

参数文件，我们在主库的pfile中已经修改，我们这里只需要该2个参数即可：

#add for standby dg
*.db_unique_name='sharon_st' 
*.log_archive_config='dg_config=(sharon_pd,sharon_st)'
*.log_archive_dest_1='location=/u01/archivelog valid_for=(all_logfiles,all_roles) db_unique_name=sharon_st'
*.log_archive_dest_2='service=sharon_pd reopen=120 lgwr async valid_for=(online_logfiles,primary_role) db_unique_name=sharon_pd'
*.log_archive_dest_state_1=enable 
*.log_archive_dest_state_2=enable 
*.standby_file_management='auto'
*.fal_server='sharon_pd'

9. 检查主库的口令文件并copy到备库

[oracle@sharon dbs]$ scp orapwdave 192.168.88.2:/u01/app/oracle/product/11.2.0/db_1/dbs

 

10. 使用RMAN备份主库（一定要掌握！）

[oracle@sharon~]$ rman target /

RMAN>RUN {

allocate channel c1 type disk;

allocate channel c2 type disk;

sql 'alter system archive log current';

BACKUP FORMAT '/u01/backup/db_%U_%T'skip inaccessible filesperset 5 DATABASE ;

sql 'alter system archive log current';

BACKUP FORMAT '/u01/backup/arch_%U_%T'skip inaccessible filesperset 5 ARCHIVELOG ALLDELETE INPUT;

backup current controlfile forstandby format='/u01/backup/control_%U';

release channel c2;

release channel c1;

}

通道可以加，通道多备份速度也可以加快

 

11. 将主库的备份文件全部copy到备库的相同位置

[oracle@sharon backup]$ scp * 192.168.88.2:/u01/backup

 

12. 在备库使用RMAN 恢复备库

备库

SQL> startup nomount

--在备库执行duplicate操作：

[oracle@sharon2 ~]$ rman target sys/oracle@dave_pdauxiliary sys/oracle@dave_st;

Duplicae数据库：

RMAN> duplicate target database for standby nofilenamecheck dorecover;

-- nofilenamecheck主备库路径相同就一定要加该参数！

 

13. 启动备库

--完成duplicate 之后，备库就是mount状态：

SQL> select open_mode from v$database;

OPEN_MODE

--------------------

MOUNTED

SQL> alter database open;

SQL> select open_mode from v$database;

OPEN_MODE

----------

READ ONLY

启动MRP，使之redo apply

SQL> alter database recover managed standbydatabase disconnect from session;

Database altered.

SQL> select open_mode from v$database;

OPEN_MODE

--------------------

READ ONLY WITH APPLY

 

14. 在主备库添加online redo log 和standby redolog

1. 主库要加standbyredo log

2. standby redolog 大小和 online redo 一样大

3. standby redolog group 比 online redo log group 多一组

SQL> col member fora50

SQL>  select group#,type, member from v$logfile;
    GROUP# TYPE    MEMBER
---------- ------- --------------------------------------------------
         3 ONLINE  /u01/app/oracle/flash_recovery_area/sharon/onlinelo
                   g/o1_mf_3_8q606gxz_.log

         2 ONLINE  /u01/app/oracle/flash_recovery_area/sharon/onlinelo
                   g/o1_mf_2_8q606c2j_.log

         1 ONLINE  /u01/app/oracle/flash_recovery_area/sharon/onlinelo
                   g/o1_mf_1_8q6066pl_.log

         4 STANDBY /u01/app/oracle/flash_recovery_area/sharon/onlinelo
                   g/o1_mf_4_8q606m0h_.log

发现恢复后的备库数据文件在FRA里，需要手动改回来，将其删除再重建。

 

删除standbylogfile，再添加

SQL>  alter database drop logfile group 4;
 alter database drop logfile group 4
*
ERROR at line 1:
ORA-01156: recovery or flashback in progress may need access to files 

–报改错是因为启动了MRP

停MRP再drop

alter database recover managed stanby database cancel;

SQL> alterdatabase drop logfile group 4;

alter databasedrop logfile group 4

*

ERRORat line 1:

ORA-00261:log 4 of thread 1 is being archived or modified

ORA-00312:online log 4 thread 1:'/home/oracle/app/flash_recovery_area/ISPACE_ST/onlinelog/o1_mf_4_8qhpmhdk_.log'

该报错说明log 4是current，先删除其他standby logfile

SQL> alter databasedrop logfile group 5;

SQL> alter databasedrop logfile group 6;

SQL> alter databasedrop logfile group 7;

添加5、6、7

SQL> ALTER DATABASEADD STANDBY LOGFILE GROUP 5 ('/u01/app/oracle/oradata/sharon/redo05.log') size50M;

SQL> ALTER DATABASEADD STANDBY LOGFILE GROUP 6 ('/u01/app/oracle/oradata/sharon/redo06.log') size50M;

SQL> ALTER DATABASEADD STANDBY LOGFILE GROUP 7 ('/u01/app/oracle/oradata/sharon/redo07.log') size50M;

用实施应用模式打开MRP，主库切换日志当log 4不是current后再drop

SQL> alter database recover managedstandby database using current logfile disconnect from session；

主库：SQL> alter system switch logfile;

SQL> alter databasedrop logfile group 4;

SQL> ALTER DATABASEADD STANDBY LOGFILE GROUP 4 ('/u01/app/oracle/oradata/sharon/redo04.log') size50M;

 

删除online redologfile，再添加

SQL> alter databasedrop logfile group 1;

alter databasedrop logfile group 1

*

ERROR at line1:

ORA-01275:Operation DROP LOGFILE is not allowed if standby file management is automatic.

SQL> alter system setstandby_file_management=MANUAL;

Systemaltered.

SQL> alter databasedrop logfile group 1;

alter databasedrop logfile group 1

*

ERROR at line1:

ORA-01623:log 1 is current log for instance sharon (thread 1) - cannot drop

ORA-00312:online log 1 thread 1: '/u01/app/oracle/flash_recovery_area/SHARON_ST/onlinelog/o1_mf_1_8qhswqjt_.log'

先删2、3

SQL> alter databasedrop logfile group 2;

Databasealtered.

SQL>  alter database drop logfile group 3;

 alter database drop logfile group 3

*

ERROR at line1:

ORA-01567:dropping log 3 would leave less than 2 log files for instance sharon (thread 1)

ORA-00312:online log 3 thread 1:'/u01/app/oracle/flash_recovery_area/SHARON_ST/onlinelog/o1_mf_3_8qhswy4h_.log'

先加group 2，再删3

SQL> alter databaseadd logfile group 2 ('/u01/app/oracle/oradata/sharon/redo02.log') size50M;

Databasealtered.

SQL> alter databasedrop logfile group 3;

Databasealtered.

SQL> alter databaseadd logfile group 3 ('/u01/app/oracle/oradata/sharon/redo03.log') size 50M;

Databasealtered.

启动MRP，主库切换日志

SQL> alter databaserecover managed standby database disconnect from session;

Databasealtered.

主库：

SQL> alter systemswitch logfile;

Systemaltered.

备库：

SQL> select * fromv$log;

 

    GROUP#   THREAD#  SEQUENCE#      BYTES BLOCKSIZE    MEMBERS ARC STATUS           FIRST_CHANGE# FIRST_TIM NEXT_CHANGE#NEXT_TIME

-------------------- ---------- ---------- ---------- ---------- --- ----------------------------- --------- ------------ ---------

         1          1         25  52428800        512          1 YES CLEARING                874946 25-APR-13       892869 25-APR-13

         2          1         26  52428800        512          1 YES CLEARING                892869 25-APR-13       892984 25-APR-13

         3          1         27  52428800        512          1 YES CLEARING_CURRENT        892984 25-APR-13            0

 

Group 1不是current，可以drop了

停MRP删group 1

SQL> alter databaserecover managed standby database cancel;

Databasealtered.

SQL> alter databasedrop logfile group 1;

alter databasedrop logfile group 1

*

ERROR at line1:

ORA-01624:log 1 needed for crash recovery of instance sharon (thread 1)

ORA-00312:online log 1 thread 1: '/u01/app/oracle/flash_recovery_area/SHARON_ST/onlinelog/o1_mf_1_8qhswqjt_.log'

SQL> alter databaseclear logfile group 1;

Databasealtered.

SQL> alter databasedrop logfile group 1;

Databasealtered.

SQL> alter databaseadd logfile group 1 ('/u01/app/oracle/oradata/sharon/redo01.log') size 50M;

Database altered.

SQL> select * fromv$logfile;

    GROUP#          STATUS  TYPE                  MEMBER                                  IS_

---------------- --------------------------------------------------------------------------------------------------------

         3         ONLINE /u01/app/oracle/oradata/sharon/redo03.log          NO

         2         ONLINE /u01/app/oracle/oradata/sharon/redo02.log          NO

         1         ONLINE /u01/app/oracle/oradata/sharon/redo01.log          NO

         4         STANDBY /u01/app/oracle/oradata/sharon/redo04.log          NO

         5         STANDBY/u01/app/oracle/oradata/sharon/redo05.log          NO

         6         STANDBY/u01/app/oracle/oradata/sharon/redo06.log          NO

         7         STANDBY /u01/app/oracle/oradata/sharon/redo07.log          NO

 

7 rowsselected.

成功！

15. 测试主库

SQL> col error for a10
SQL> col dest_name for a20
SQL> select DEST_NAME,STATUS,PROCESS,ERROR,TRANSMIT_MODE from v$archive_dest WHERE TARGET='STANDBY';

DEST_NAME            STATUS    PROCESS   ERROR      TRANSMIT_MOD

----------------------------- ---------- ---------- ------------

LOG_ARCHIVE_DEST_2   VALID    LGWR                  ASYNCHRONOUS

SQL> SELECT PROCESS,STATUS, THREAD#, SEQUENCE#, BLOCK#, BLOCKS FROM V$MANAGED_STANDBY;

PROCESS   STATUS          THREAD#  SEQUENCE#    BLOCK#     BLOCKS

--------------------- ---------- ---------- ---------- ----------

ARCH      CLOSING               1        43          1          6

ARCH      CLOSING               1         23          1       2574

ARCH      CLOSING               1         44          1          4

ARCH      CLOSING               1         45          1          2

LNS       WRITING               1         46        156          1

SQL>  select sequence#,applied from v$archived_log;

 SEQUENCE# APPLIED
---------- ---------
        43 YES
        44 YES
        45 YES

主库：

SQL> selectmax(sequence#) from v$archived_log;

SQL> altersystem switch logfile;

SQL> altersystem switch logfile;

SQL> selectmax(sequence#) from v$archived_log;

SQL> SELECTPROCESS, STATUS, THREAD#, SEQUENCE#, BLOCK#, BLOCKS FROM V$MANAGED_STANDBY;     --速度要快，就可以看到LNS进程了！

SQL>  select sequence#,applied from v$archived_log;

 

有问题时如何考虑？

1. 检查日志是否传过来

2. 为什么不apply

1） mrp 进程没有启动

        ps-ef |grep mrp

2) alert  log