oracle data guard的搭建

准备工作：

主库装好oracle软件，并建库，备库只装软件

修改hosts文件

关闭防火墙

 

主库：

开启force强制日志模式：

SQL>

alter database force logging;

select force_logging from v$database;

 

开启归档

主备库都要建立归档日志目录：mkdir  /u01/archivelog

SQL>

shutdown immediate;   

startup mount;

alter database archivelog;

alter database open;

alter system set log_archive_dest_1='location=/u01/archivelog/';

archive log list ;

 

查询主库状态：

SQL>

select instance_name,status from v$instance;

 

查询主库参数：

SQL>

show parameter name

 

修改主库的tnsnames.ora文件：

#

cd $ORACLE_HOME/network/admin

cat tnsnames.ora

 

DG1 =

  (DESCRIPTION =

    (ADDRESS = (PROTOCOL = TCP)(HOST = dg1)(PORT = 1521))

    (CONNECT_DATA =

      (SERVER = DEDICATED)

      (SERVICE_NAME = DG1)

    )

  )

DG2 =

  (DESCRIPTION =

    (ADDRESS = (PROTOCOL = TCP)(HOST = dg2)(PORT = 1521))

    (CONNECT_DATA =

      (SERVER = DEDICATED)

      (SERVICE_NAME = DG2)

      (UR = A)

    )

  )

 

将tnsname.ora传到备库

#

scp tnsname.ora dg2:$ORACLE_HOME/network/admin/

 

 

修改主库密码文件,并传到备库上：

#

cd $ORACLE_HOME/dbs

orapwd file=orapwdg password=redhat entries=3 force=y;

 

查询sid

env|grep ora

 

创建alert日志的软连接：

ln -s /u01/app/oracle/diag/rdbms/dg1/dg/trace/alert_dg.log /u01/alert.log

tailf /u01/alert.log

 

查询主库参数：

SQL> show parameter name；

 

修改主库参数，并传到备库:

alter system set log_archive_config='DG_CONFIG=(dg1,dg2)';

alter system set log_archive_dest_1="LOCATION=/u01/app/oracle/archivelog/ LGWR VALID_FOR=(ALL_LOGFILES,ALL_ROLES) DB_UNIQUE_NAME=DG1";

alter system set log_archive_dest_2="SERVICE=DG2 ASYNC VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=DG2";

alter system set log_archive_dest_state_1=enable scope=both;

alter system set log_archive_dest_state_2=defer scope=both;

alter system set standby_file_management='AUTO';

alter system set fal_server=dg2 scope=both;

alter system set db_file_name_convert="/u01/app/oracle/oradata/dg1","/u01/app/oracle/oradata/dg1" scope=spfile;

alter system set log_file_name_convert="/u01/app/oracle/oradata/dg1","/u01/app/oracle/oradata/dg1" scope=spfile;

 

生成参数文件，并传给备库：

SQL>

create  pfile  from  spfile;

#

scp  $ORACLE_HOME/dbs/initdg.ora  dg2:$ORACLE_HOME/dbs/

 

全量备份主库：

#

主备库都要建立备份目录mkdir /u01/rmanbak/

rman  target  /

 

run

{

allocate channel c0 device type disk;

allocate channel c1 device type disk;

CONFIGURE CONTROLFILE AUTOBACKUP ON;

CONFIGURE CONTROLFILE AUTOBACKUP FORMAT FOR DEVICE TYPE DISK TO '/u01/rmanbak/%F';

backup database format '/u01/rmanbak/ora11g_full_db_%d_%T_%u.bak';

BACKUP ARCHIVELOG ALL FORMAT '/u01/rmanbak/ora11g_arc_%s_%p_%t.bak';

}

 

查询是否有控制文件的备份：

list backup of controfile;

 

将备份文件传输的备库：

scp  /u01/rmanbak/* dg2:/u01/rmanbak/

 

 

备库：

 

编辑备库参数文件：

此处应添加db_unique_name=DG2

其他几个按需要改。

 

备库启动监听：

lsnrctl   start

 

链接alert日志：

ln -s /u01/app/oracle/diag/rdbms/dg1/dg/trace/alert_dg.log /u01/alert.log

 

开启备库到nomount

SQL>

startup nomout;

!lsnrctl status

 

主库操作：

rman   target  /

connect auxiliary sys/redhat@dg2

duplicate target database for standby nofilenamecheck;

过程中持续看alert日志，若报错，则是initdg.ora文件参数设置问题，修改后，重新开始，若还不行，则用下面的命令

duplicate target database for standby nofilenamecheck;

 

添加主备库的standby日志

alter database add standby logfile group  11 '/u01/app/oracle/oradata/dg1/standby11.log' size 300M;

alter database add standby logfile group  12 '/u01/app/oracle/oradata/dg1/standby12.log' size 300M;

alter database add standby logfile group  13 '/u01/app/oracle/oradata/dg1/standby13.log' size 300M;

alter database add standby logfile group  14 '/u01/app/oracle/oradata/dg1/standby14.log' size 300M;

alter database add standby logfile group  15 '/u01/app/oracle/oradata/dg1/standby15.log' size 300M;

查询：

SQL> select member from v$logfile ;

查询主备库:

select database_role from v$database;

 

 

主库开启日志：

SQL>

show parameter log_archive_dest_state_;

alter system set log_archive_dest_state_2=enable scope=both;

 

开启备库

SQL>

alter database recover managed standby database using current logfile disconnect from session;

alter database recover managed standby database cancel;

alter database open;

alter database recover managed standby database using current logfile disconnect from session;

 

测试：

主库添加

SQL>

create table  t1 (id int);

insert into t1 values(2);

commit;

归档一份日志

alter system switch logfile;