Content Preview: rss
946 days ago
添加一个新的复制表 源数据库 添加附加日志 ALTER TABLE "REP"."C" ADD SUPPLEMENTAL LOG DATA (PRIMARY KEY, FOREIGN KEY, UNIQUE INDEX) COLUMNS; 实例化准备 DBMS_CAPTURE_ADM.PREPARE_TABLE_INSTANTIATION( table_name => 'rep.c', supplemental_logging => 'keys'); 添加捕捉规则 dbms_streams_adm.add_table_rules ( table_name => '"REP"."C"', streams_type => 'CAPTURE', streams_name => '"TEST$CAP"', queue_name => '"STREAMADMIN"."TEST$CAPQ"', include_dml => TRUE, include_ddl => TRUE, include_tagged_lcr => TRUE, source_database => 'TEST.REGRESS.RDBMS.DEV.US.ORACLE.COM', inclusion_rule => TRUE); 添加传送规则 dbms_streams_adm.add_table_propagation_rules ( table_name => '"REP"."C"', streams_name => '', ...
946 days ago
创建单源流复制数据库 源数据库 创建流复制管理员 CREATE USER STREAMADMIN IDENTIFIED BY STREAMADMIN DEFAULT TABLESPACE USERS" TEMPORARY TABLESPACE "TEMP" ACCOUNT UNLOCK; GRANT CONNENT TO STREAMADMIN ; GRANT DBA TO STREAMADMIN ; 创建数据库链接 CREATE DATABASE LINK "DEST" -- 数据库链接名，必须在 netca 中配置过 CONNECT TO "STREAMADMIN" -- 目的数据库时复制管理员 IDENTIFIED BY "STREAMADMIN" -- 目的数据库时复制管理员密码 USING 'DEST' ; -- 目的数据库名，必须在 netca 中配置过 创建捕捉队列 dbms_streams_adm.set_up_queue ( queue_table => '"STREAMADMIN"."TEST$CAPQT"', storage_clause => NULL, queue_name => '"STREAMADMIN"."TEST$CAPQ"', queue_user => ''); 对要复制的表实例化准备 ...
1004 days ago
数据缓冲区命中率: (通常>90%，否则请调整sga_target或db_cache_size) select 100 - ( pr - x) / (cg + dg - x)*100 "db_buffer hit radio(%)" from (select value pr from v$sysstat where name ='physical reads'), (select sum(value) x from v$sysstat where name ='physical reads direct' or name ='physical reads direct (lob)'), (select value cg from v$sysstat where name ='consistent gets'), (select value dg from v$sysstat where name = 'db block gets') / 共享池的命中率: (通常>95%,甚至99%,否则请调整sga_target或share_pool_size) select sum(pinhits-reloads)/sum(pins)*100 "Share pool hit radio(%)" from v$librarycache / 数据字典命中率: select (1 - (sum(getmisses) / sum(gets))) * 100 "Data Dictionary Hit Ratio(%)" from v$rowcache; / 排序区: (disk与memory的比不能过高，否则调整sort_area_size select name,value from v$sysstat where name like '%sort%' / log_buffer: (通常redo entries/redo buffer allocation retries > 100否则调整log_buffer): select name,value from v$sysstat where name in('redo entries','redo buffer ...
1166 days ago
添加节点: 1.使用$ORA_CRS_HOME/oui/bin/addNode.sh添加CRS 2.使用DBCA的实例管理添加数据库 修改vip及vip的机器名: 1.停掉所有节点数据库,ASM. 2.停掉所有节点应用(srvctl stop nodeapps -n <node_name > ) 3.srvctl remove nodeapps -n <node_name > [-f] 4.删除侦听配置(netca) 5.修改/etc/hosts 6.使用vipca配置 7.创建侦听(netca) 仅修改vip: 1.停掉所有节点数据库,ASM. 2.停掉所有节点应用(srvctl stop nodeapps -n <node_name > ) 2.停掉所有节点crs(/etc/init.d/init.crs stop) 3.修改所有节点操作系统ip,/etc/hosts文件 4.起动所有节点crs(/etc/init.d/init.crs start) 5.停掉所有节点应用(srvctl stop nodeapps -n <node_name > ) 6.修改所有节点vip(srvctl modify nodeapps -A 192.168.55.21/255.255.255.0/eth0) 7.启动所有节点应用(srvctl start nodeapps -n ora1) 8.启动所有节点数据库,ASM racgvip 修改内外网连接 crsctl crs_stat crs_start,crs_stop 控制查看crs状态 srvctl 打开跟踪:export SRVM_TRACE=true 查看配置:srvctl config <object> ... 解决节点应用不能启动脚本: srvctl stop nodeapps -n `hostname` for app in `crs_stat grep NAME awk -F= '{print $2}'` do if [ `crs_stat $app ...
