ADG搭建，RAC--＞单节点，搭建过程及故障描述

主库:ecology1        10.1.1.95      Centos 7.9

         ecology2         10.1.1.96     Centos 7.9

备库:ecologydg        192.168.16.3   Centos 7.9

1.首先，主库开启归档（两节点需要重启）,force login

两节点:

SQL>shutdown immediate;

SQL>startup mount 

SQL>alter database archivelog        #开启归档

SQL>alter database force logging;    #开启force login

SQL>alter system set log_archive_dest_1='location=+DATA' sid='*' scope=spfile;#设置本地归档

SQL>startup;

SQL> select log_mode, force_logging from v$database;   #查看数据库模式

LOG_MODE			     FORCE_LOG
------------------------------------ ---------
ARCHIVELOG			     YES

2.主库配置dataguard 参数

SQL>alter system set log_archive_config='dg_config=(ecology,ecologydg)' sid='*';#打开dataguard配置
SQL>alter system set log_archive_dest_3='service=ecologydg valid_for=(online_logfiles,primary_role) db_unique_name=ecologydg' sid='*';   #设置远程归档
scp $ORACLE_HOME/dbs/orapw$ORACLE_SID 192.168.16.3:$ORACLE_HOME/dbs/orapwecologydg   #拷贝主库密码文件至备库

3.备库配置dataguard参数。并创建相应目录

直接拷贝init.ora为initecologydg.ora  编辑

*.audit_file_dest='/u01/app/oracle/admin/ecologydg/adump'
*.audit_trail='db'
*.compatible='11.2.0.4.0'
*.control_files='/u01/app/oracle/oradata/ecologydg/controlfile/control01.ctl'
*.db_block_size=8192
*.db_domain=''
*.db_name='ecology'
*.diagnostic_dest='/u01/app/oracle'
*.fal_client='ecologydg'
*.fal_server='ecology'
*.log_archive_config='dg_config=(ecology,ecologydg)'
*.log_archive_dest_2='location=/oracle/dataguard/ecologydg/archivelog/ valid_for=(standby_logfiles,standby_role) db_unique_name=ecologydg'
*.memory_target=4G
*.open_cursors=300
*.processes=500
*.remote_login_passwordfile='exclusive'
*.undo_tablespace='UNDOTBS1'
standby_file_management=auto
db_unique_name='ecologydg'
db_file_name_convert='+DATA/ecology/datafile/','/oracle/dataguard/ecologydg/datafile/','+DATA/ecology/tempfile/','/u01/app/oracle/oradata/ecologydg/tempfile/'
log_file_name_convert='+DATA/ecology/onlinelog/','/oracle/dataguard/ecologydg/onlinelog/'

mkdir -p /u01/app/oracle/admin/ecologydg/adump
mkdir -p /u01/app/oracle/oradata/ecologydg/controlfile
mkdir -p /u01/app/oracle/oradata/ecologydg/tempfile
mkdir -p /oracle/dataguard/ecologydg/onlinelog
mkdir -p /oracle/dataguard/ecologydg/datafile
mkdir -p /oracle/dataguard/ecologydg/archivelog

4.主库配置tnsname,备库修改为静态监听。

主库两个节点都配置

vim $ORALCE_HOME/network/admin/tnsname.ora

ecologydg=
  (DESCRIPTION =
    (ADDRESS_LIST =
      (ADDRESS = (PROTOCOL = TCP)(HOST = 192.168.16.3)(PORT = 1521))
    )
    (CONNECT_DATA =
      (SERVER = DEDICATED)
      (SERVICE_NAME = ecologydg)
    )
  )

备库

vim $ORALCE_HOME/network/admin/listener.ora

SID_LIST_LISTENER =
  (SID_LIST =
    (SID_DESC =
      (GLOBAL_DBNAME = ecologydg)
      (ORACLE_HOME = /u01/app/oracle/product/11.2.0/db_1)
      (SID_NAME = ecologydg)
    )
  )

LISTENER =
  (DESCRIPTION =
    (ADDRESS = (PROTOCOL = TCP)(HOST = 192.168.16.3)(PORT = 1521))
  )

lsnrctl status     #状态为unknow 开启成功，静态监听的作用是在duplicate的时候备库状态为nomount依然可以连接。

5.duplicate 复制主库,并创建standby_log

sqlplus sys/password@ecologydg as sysdba  #测试主库至备库的连通性。

登录rman ,开始duplicate，主库备库都可以执行

rman target / auxiliary sys/password@ecologydg log=/opt/dublicate230905_log append <<EOF
run {
duplicate target database for standby from active database;
}
EOF

mkdir -p /oracle/dataguard/ecologydg/standbylog/

#数量比节点一节点二加起来多两组
alter database add standby logfile thread 1 group 21 '/oracle/dataguard/ecologydg/standbylog/std01.log' size 50M;
alter database add standby logfile thread 1 group 22 '/oracle/dataguard/ecologydg/standbylog/std02.log' size 50M;
alter database add standby logfile thread 1 group 23 '/oracle/dataguard/ecologydg/standbylog/std03.log' size 50M;
alter database add standby logfile thread 2 group 24 '/oracle/dataguard/ecologydg/standbylog/std04.log' size 50M;
alter database add standby logfile thread 2 group 25 '/oracle/dataguard/ecologydg/standbylog/std05.log' size 50M;
alter database add standby logfile thread 2 group 26 '/oracle/dataguard/ecologydg/standbylog/std06.log' size 50M;

6.备库应用日志

alter database recover managed standby database disconnect from session; #先同步日志

alter database recover managed standby database cancel;  #取消应用

alter database open;  #起库

alter database recover managed standby database using current logfile disconnect from session;
#开启实时应用

---查看dg信息

SELECT MESSAGE FROM V$DATAGUARD_STATUS;
 

---查看进程状态
select process,client_process,sequence#,status,THREAD# from v$managed_standby;

#redo情况

select thread#,max(sequence#) from v$log group by thread# order by 1;
 

#日志应用情况

select thread#,max(sequence#) from v$archived_log where applied='YES' group by thread#;
 

备库切换主库下次再写