DG搭建删除 - 勇敢DBA不怕困难

#备库删除命令方法一： shutdown immediate startup mount alter system enable restricted session; —只有restricted session 权限的用户可以登录数据库，已存在的session不会中断 drop database;

方法二：（asmcmd删除数据文件，删除前先关闭数据库） shutdown immediately arch_dg rm -rf ARCHIVELOG CONTROLFILE ONINELOG data_dg rm -rf CONTROLFILE DATAFILE ONINELOG TEMPFILE

参数文件要放到asm的话，要建到asm中，并在 $ORACLE_HOME/dbs里创建init$ ORACLE_SID.ora,设置spfile参数，删除spfile$ORACLE_SID.ora及spfile.ora文件

#修改数据库参数主库参数： show parmaeter unique ALTER SYSTEM SET LOG_ARCHIVE_CONFIG=‘DG_CONFIG=(ogg,dgogg)’ SCOPE=both; —主备数据库唯一名

ALTER SYSTEM SET LOG_ARCHIVE_DEST_2=‘SERVICE=DGOGGdg LGWR ASYNC VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=dgogg’ SCOPE=both; 服务名=备库的tns 备库唯一名 location指定本地路径，service为dataguard备库的service name 可通过查询V$ARCHIVE_DEST 视图来查看具体的属性设置该参数必须和LOG_ARCHIVE_DEST_STATE_n一一对应

ALTER SYSTEM SET LOG_ARCHIVE_DEST_STATE_1 = { enable | defer | alternate } SCOPE=both; 取值范围 defer alternate 默认值 enable 指定对应的归档日志路径是生效的，这是默认值 defer 指定对应的归档日志路径是暂时失效的，使用需重新开启(enable) alternate 指定对应的归档日志路径是备用的，当另外路径失效时启用该路径设置LOG_ARCHIVE_DEST_n 时相应的LOG_ARCHIVE_DEST_STATE_n也需设置通过 V$ARCHIVE_DES视图查看该属性值

ALTER SYSTEM SET FAL_CLIENT=OGGdg SID='' SCOPE=both; —主库tns名 ALTER SYSTEM SET FAL_SERVER=DGOGGdg SID='' SCOPE=both; —备库tns名，可以填多个

ALTER SYSTEM SET STANDBY_FILE_MANAGEMENT=‘AUTO’ SID=’*’ SCOPE=both; —主库创建新的数据文件，备库自动创建该文件

磁带库起库 vim rman_recover.sh #!/usr/bin/ksh echo -------------------------begin--------------------------- date rman target sys/123456@OGGdg auxiliary sys/123456@DGOGG_rman <<EOF run { allocate channel prmy1 device type disk connect ‘sys/123456@OGGdg’; —并行 allocate channel prmy2 device type disk connect ‘sys/123456@OGGdg’; … allocate channel prmy10 device type disk connect ‘sys/123456@OGGdg’; allocate auxiliary channel stdy1 device type disk; —并行 allocate auxiliary channel stdy2 device type disk; … allocate auxiliary channel stdy10 device type disk; duplicate target database for standby from active database nofilenamecheck; sql channel prmy1 “alter system archive log current”; sql channel stdy1 “alter database recover managed standby database disconnect”; release channel prmy1; release channel prmy2; … release channel prmy10; release channel stdy1; release channel stdy2; … release channel stdy10; } exit; EOF sleep 90 sqlplus / as sysdba <<EOF alter database recover managed standby database cancel; alter database add standby logfile thread 2 group 31 ‘+DATA_DG’ size 1024m; alter database add standby logfile thread 2 group n ‘+DATA_DG’ size 1024m; alter database open; ALTER DATABASE RECOVER MANAGED STANDBY DATABASE USING CURRENT LOGFILE DISCONNECT; exit EOF date echo “-----------------------------END---------------------------”

执行脚本 chmod +x rman_recover.sh nohup rman_recover.sh > rman_recover.log & tail -2000f recover.log