主库宕机后，从库切换为主库，重新搭建DG过程如下，

1、修改primary端初始化参数文件（不用修改，目前就是这样的配置）
alter system set log_archive_config = 'DG_CONFIG=(zzdb,zzdb_dg)' scope=both;
alter system set log_archive_dest_1 = 'LOCATION=/oracle/archive VALID_FOR=(ALL_LOGFILES,ALL_ROLES) DB_UNIQUE_NAME=zzdb_dg' scope=both;
alter system set log_archive_dest_2 = 'SERVICE=zzdb LGWR ASYNC VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=zzdb' scope=both;
alter system set log_archive_dest_state_1 = enable;
alter system set log_archive_dest_state_2 = enable;
alter system set fal_server=zzdb scope=both;
alter system set fal_client=zzdb_dg scope=both;
alter system set standby_file_management=AUTO scope=both;
alter system set db_file_name_convert='/oracle/product/oradata/zzdb','/oracle/product/oradata/zzdb' scope=spfile;
alter system set log_file_name_convert='/oracle/product/oradata/zzdb','/oracle/product/oradata/zzdb' scope=spfile;
2、现primary端pfile参数文件和密码文件，并且拷贝到standby节点（原primary节点）段相应位置
SQL> create pfile from spfile;
cd $ORACLE_HOME/dbs
scp initzzdb.ora orapwzzdb ora11-node1:$ORACLE_HOME/dbs
cd $ORACLE_BASE
scp -r admin/ diag/ ora11-node1:$ORACLE_BASE
3、修改standby（原primary节点）的监听文件及初始化参数文件
--复制现primary节点的监听文件
--启动监听
[oracle@ora11-node1 dbs]$ lsnrctl start
--修改参数文件
[oracle@ora11-node1 ~]$ sqlplus / as sysdba
SQL> startup nomount;
SQL> create spfile from pfile='$ORACLE_HOME/dbs/initzzdb.ora';
SQL> shutdown abort;
SQL> startup nomount;
SQL>
alter system set db_unique_name=zzdb scope=spfile;
alter system set log_archive_config='DG_CONFIG=(zzdb,zzdb_dg)' scope=both;
alter system set log_archive_dest_1 = 'LOCATION=/oracle/archive VALID_FOR=(ALL_LOGFILES,ALL_ROLES) DB_UNIQUE_NAME=zzdb' scope=spfile;
alter system set log_archive_dest_2 = 'SERVICE=zzdb_dg LGWR ASYNC VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=zzdb_dg' scope=spfile;
alter system set fal_server=zzdb_dg scope=spfile;
alter system set fal_client=zzdb scope=spfile;
SQL> shutdown abort;
SQL> startup nomount;
4、在primary端通过Rman创建备库,在 ora11-node2 上执行如下命令
#primary备份数据库
mkdir -p /oracle/db_backup
rman target / <<!
run {
CONFIGURE DEVICE TYPE DISK PARALLELISM 3;
backup database format '/oracle/db_backup/%d_%s_%p_%t.%T';
}
!
# primary生成备库控制文件
rman target / <<!
run {
allocate channel c1 device type disk format '/oracle/db_backup/CON_%U';
backup current controlfile for standby;
}
!
# cp备份文件和控制文件到standby
cd /oracle/db_backup
scp * ora11-node1:/oracle/db_backup
#standby节点，恢复控制文件
rman target / <<!
run {
restore controlfile from '/oracle/db_backup/CON_0bqq71jt_1_1';
alter database mount;
}
!
#standby节点，通过rman进行恢复
rman target / <<!
CONFIGURE DEVICE TYPE DISK PARALLELISM 3;
restore database;
!
#至此已经恢复完成
5、在 standby 端查看standby日志
SQL> SELECT GROUP#,THREAD#,SEQUENCE#,ARCHIVED,STATUS FROM V$STANDBY_LOG;
若没有添加：
SQL> alter database add standby logfile
group 4 ('/oracle/product/oradata/zzdb/zzdb_dgredo04.log') size 50m,
group 5 ('/oracle/product/oradata/zzdb/zzdb_dgredo05.log') size 50m,
group 6 ('/oracle/product/oradata/zzdb/zzdb_dgredo06.log') size 50m,
group 7 ('/oracle/product/oradata/zzdb/zzdb_dgredo07.log') size 50m;
6、在standby端开启实时日志应用
SQL> alter database recover managed standby database using current logfile disconnect from session;
recover managed standby database disconnect from session;
startup;

ITPub博客

ora11_node_dg（4）主库宕机后，从库升为主库，重新搭建DG

最新文章