ORACLE Physical Standby DG搭建
主庫：
一：強制force logging: alter database force logging;
二：開啟主庫的歸檔模式
三：主庫添加standby redo log，比redo日志組多一組：
ALTER DATABASE ADD STANDBY LOGFILE GROUP 4 ('/data/app/oracle/oradata/PROD/redo04_1.log','/data/app/oracle/oradata/PROD/redo04_2.log') size 1024M;
ALTER DATABASE ADD STANDBY LOGFILE GROUP 5 ('/data/app/oracle/oradata/PROD/redo05_1.log','/data/app/oracle/oradata/PROD/redo05_2.log') size 1024M;
ALTER DATABASE ADD STANDBY LOGFILE GROUP 6 ('/data/app/oracle/oradata/PROD/redo06_1.log','/data/app/oracle/oradata/PROD/redo06_2.log') size 1024M;
ALTER DATABASE ADD STANDBY LOGFILE GROUP 7 ('/data/app/oracle/oradata/PROD/redo07_1.log','/data/app/oracle/oradata/PROD/redo07_2.log') size 1024M;
select * from v$standby_log;
四：創(chuàng)建pfile文件
主庫初始化參數(shù)的配置：注意：內(nèi)存分配、數(shù)據(jù)目錄、日志目錄、DBname主備相同，DB_UNIQUE_NAME主備不同，11g動態(tài)注冊的監(jiān)聽服務(wù)名、數(shù)據(jù)庫的service_names與DB_UNIQUE_NAME一致
PROD.__data_transfer_cache_size=0
PROD.__db_cache_size=2415919104
PROD.__java_pool_size=16777216
PROD.__large_pool_size=33554432
PROD.__oracle_base='/data/oracle/product'#ORACLE_BASE set from environment
PROD.__pga_aggregate_target=1241513984
PROD.__sga_target=6861881344
PROD.__shared_io_pool_size=251658240
PROD.__shared_pool_size=4093640704
PROD.__streams_pool_size=33554432
*.audit_file_dest='/data/app/oracle/admin/PROD/adump'
*.audit_trail='db'
*.compatible='12.1.0.2.0'
*.control_files='/data/app/oracle/oradata/PROD/control01.ctl','/data/app/oracle/fast_recovery_area/PROD/control02.ctl'
*.db_block_size=8192
*.db_domain=''
*.db_name='PROD'
*.db_recovery_file_dest='/data/app/oracle/fast_recovery_area'
*.db_recovery_file_dest_size=4560m
*.diagnostic_dest='/data/app/oracle'
*.dispatchers='(PROTOCOL=TCP) (SERVICE=PRODXDB)'
#*.log_archive_dest_1='location=/data/app/oracle/archivelog'
*.open_cursors=300
*.pga_aggregate_target=1512m
*.processes=5000
*.local_listener='LISTENER_PROD'
*.remote_login_passwordfile='EXCLUSIVE'
*.sga_target=7536m
*.undo_tablespace='UNDOTBS1'
DB_UNIQUE_NAME=PROD
LOG_ARCHIVE_CONFIG='DG_CONFIG=(prod,proddg)'
LOG_ARCHIVE_DEST_1= 'LOCATION=/data/app/oracle/archivelog VALID_FOR=(ALL_LOGFILES,ALL_ROLES) DB_UNIQUE_NAME=prod'
LOG_ARCHIVE_DEST_2= 'SERVICE=proddg ASYNC VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=proddg'
REMOTE_LOGIN_PASSWORDFILE=EXCLUSIVE
LOG_ARCHIVE_FORMAT=%t_%s_%r.arc
LOG_ARCHIVE_DEST_1='enable'
LOG_ARCHIVE_DEST_2='enable'
FAL_CLIENT=prod
FAL_SERVER=proddg
DB_FILE_NAME_CONVERT='/data/app/oracle/oradata/PROD/','/data/app/oracle/oradata/PROD/'
LOG_FILE_NAME_CONVERT='/data/app/oracle/oradata/PROD/','/data/app/oracle/oradata/PROD/'
STANDBY_FILE_MANAGEMENT=AUTO
五：在主備庫創(chuàng)建靜態(tài)listener和tnsnames.ora，主庫創(chuàng)建口令文件，發(fā)送到備庫
LISTENER=
(DESCRIPTION=
(ADDRESS_LIST=
(ADDRESS=(PROTOCOL=tcp)(HOST= 10.26.208.131)(PORT=1521))
(ADDRESS=(PROTOCOL=ipc)(KEY=extproc))))
SID_LIST_LISTENER=
(SID_LIST=
(SID_DESC=
(GLOBAL_DBNAME=prod)
(ORACLE_HOME=/u01/app/oracle/product/11.2.0/db_1)
(SID_NAME=prod)
)
)
cat tnsnames.ora
prod=
(DESCRIPTION=
(ADDRESS=(PROTOCOL=tcp)(HOST=10.26.208.130)(PORT=1521))
(CONNECT_DATA=(SERVICE_NAME=prod))
)
proddg=
(DESCRIPTION=
(ADDRESS=(PROTOCOL=tcp)(HOST=10.26.208.131)(PORT=1521))
(CONNECT_DATA=(SERVICE_NAME=prod))
)
口令文件：orapwd file=orapwinstance_name（實例名） password= entries=10 force=y
在主庫創(chuàng)建口令文件orapwpdb，然后scp到備庫，在備庫修改名為orapwsdb
六：在備庫修改初始化參數(shù)文件pfile
七：對主庫進行備份，發(fā)送至備庫。
在主備庫，創(chuàng)建/backup備份目錄
對主庫進行備份：
run{
allocate channel c1 type disk;
allocate channel c2 type disk;
sql 'alter system archive log current';
backup format '/backup/db_%U_%T' skip inaccessible filesperset 5 database;
sql 'alter system archive log current';
backup format '/backup/db_%U_%T' skip inaccessible filesperset 5 archivelog all delete input;
backup current controlfile for standby format='/backup/control_%U';
release channel c2;
release channel c1;
}
八：主備機器配置雙向SSH互信，將主庫的備份復(fù)制到備庫的/backup，相同目錄，如果目錄不相同，則使用catalog start with '備份所在的路徑' 或者 catalog 例如：catalog backuppiece '/node2/database/backup/o1_mf_annnn_TAG20070213T002925_2x21m6ty_.bkp';
scp /backup/* 192.168.9.2:/backup/
九：開啟備庫至nomount,進行恢復(fù)
SQL>startup nomount
在備庫進行恢復(fù)
RMAN>rman target sys/sys@prod auxiliary sys/sys@proddg
RMAN> duplicate target database for standby nofilenamecheck from active database;
十：打開備份，查詢狀態(tài)，可能需要應(yīng)用歸檔進行恢復(fù)
重建備庫的standby redo log（可選）
最后，驗證主庫
select dest_name,status,process,error,transmit_mode from v$archive_dest where target='STANDBY';
主備切換驗證
問題處理：一主五從
ORA-16198: Timeout incurred on internal channel during remote archival
SQL>select dest_name,status,process,error,transmit_mode from v$archive_dest where target='STANDBY';
DEST_NAME STATUS PROCESS ERROR TRANSMIT_MOD
-------------------- --------- ---------- ----------------------------------------------------------------- ------------
LOG_ARCHIVE_DEST_2 ERROR LGWR ORA-16198: Timeout incurred on internal channel during remote ASYNCHRONOUS archival
LOG_ARCHIVE_DEST_3 VALID LGWR PARALLELSYNC
LOG_ARCHIVE_DEST_4 VALID LGWR ASYNCHRONOUS
LOG_ARCHIVE_DEST_5 VALID LGWR PARALLELSYNC
LOG_ARCHIVE_DEST_6 VALID LGWR ASYNCHRONOUS
SQL> alter system set log_archive_dest_2='service=hbhsdb LGWR ASYNC NET_TIMEOUT=40 valid_for=(ONLINE_LOGFILES,PRIMARY_ROLE) db_unique_name=prod';
SQL>select dest_name,status,process,error,transmit_mode from v$archive_dest where target='STANDBY';
DEST_NAME STATUS PROCESS ERROR TRANSMIT_MOD
-------------------- --------- ---------- ----------------------------------------------------------------- ------------
LOG_ARCHIVE_DEST_2 VALID LGWR ASYNCHRONOUS
LOG_ARCHIVE_DEST_3 VALID LGWR PARALLELSYNC
LOG_ARCHIVE_DEST_4 VALID LGWR ASYNCHRONOUS
LOG_ARCHIVE_DEST_5 VALID LGWR PARALLELSYNC
LOG_ARCHIVE_DEST_6 VALID LGWR ASYNCHRONOUS
ORA-16024: parameter LOG_ARCHIVE_DEST_1 cannot be parsed
在備庫啟動到nomount，報錯ORA-16024
解決方式：在pfile文件注釋掉下列兩個參數(shù)
LOG_ARCHIVE_DEST_1='enable'
LOG_ARCHIVE_DEST_2='enable'
總結(jié)
以上是生活随笔為你收集整理的ORACLE Physical Standby DG搭建的全部內(nèi)容，希望文章能夠幫你解決所遇到的問題。
- 上一篇： bzoj 5092 分割序列 —— 高维
- 下一篇： vue2和vue3区别