一，软件安装
源端和目标端均执行（只要修改相应的目录）
1.上传软件，放到ogg的安装目录，并解压
mkdir /home/oracle/ogg
unzip ogg112101_fbo_ggs_Linux_x64_ora11g_64bit.zip
tar -xvf fbo_ggs_Linux_x64_ora11g_64bit.tar
2.设置环境变量
export GGATE_HOME=/home/oracle/ogg
export LIBPATH=GGATEHOME:ORACLE_HOME/lib
. .bash_profile
3.进入软件安装目录
cd /home/oracle/ogg/
./ggsci
create subdirs
4. 在ORACLE中创建ogg专属用户
源端数据库：
sqlplus /nolog
conn / as sysdba
create user ogg identified by ogg;
grant connect,resource to ogg;
grant unlimited tablespace to ogg;
grant execute on utl_file to ogg;
grant SELECT ANY DICTIONARY, SELECT ANY TABLE to ogg;
grant ALTER ANY TABLE to ogg;
grant FLASHBACK ANY TABLE to ogg;
grant EXECUTE on DBMS_FLASHBACK to ogg;
grant executeon utl_file to ogg;
grant alter any table to ogg;
grant flashback any table to ogg;
grant execute on DBMS_FLASHBACK to ogg;
目标端数据库：
sqlplus /nolog
conn / as sysdba
create user ogg identified by ogg;
grant connect,resource,unlimited tablespace to ogg;
grant execute on utl_file to ogg;
grant select any dictionary,select any table to ogg;
grant alter any table to ogg;
grant flashback any table to ogg;
grant execute on DBMS_FLASHBACK to ogg;
grant insert any table to ogg;
grant delete any table to ogg;
grant update any table to ogg;
5.创建测试表
(下面的例子以oracle自带的hr用户为例)
源端测试表：create table test as select object_id from dba_objects where rownum<101; ---插入100条数据
目标端测试表：create table test as select object_id from dba_objects where 1=2; --不插入数据
二，源端设置
1.开启ORACLE归档模式，设置日志模式（归档模式、附加日志、强制日志）
--查看
select db.LOG_MODE, db.SUPPLEMENTAL_LOG_DATA_MIN, db.FORCE_LOGGING from v$database db
--修改
--1）archivelog
shutdown immediate
startup mount
alter database archivelog;
alter database open;
--2)force logging
alter database force logging;
--3)supplemental log data
alter database add supplemental log data;
2. 配置管理进程
EDIT PARAMS MGR
PORT 7809
启动
START MGR
3. 添加抽取进程：
GGSCI (db11) 31> view param lhremd1
EXTRACT lhremd1
setenv(NLS_LANG=AMERICAN_AMERICA.ZHS16GBK)
USERID ogg, PASSWORD ogg
DBOPTIONS ALLOWNOLOGGING
EXTTRAIL ./dirdat/l1
TABLE hr.*;
添加：
GGSCI (sourceserver) > ADD EXTRACT LHREMD1, TRANLOG, BEGIN NOW
GGSCI (sourceserver) > ADD EXTTRAIL dirdat/l1, EXTRACT LHREMD1, MEGABYTES 100
启动
start ext lhremd1
4. 打开表的补全日志：
GGSCI (dsgdata) 52>add trandata hr.employees
5. 添加投递进程：
GGSCI (db11) 32> view param phremd1 Extract PHREMD1 ------------------------------------------------------------------- -- Data Pump extract for HR schema ------------------------------------------------------------------- PassThru RmtHost 192.168.1.3, MgrPort 7809 RmtTrail dirdat/l2 Table HR.* ;
-----说明：hr用户下所有表中执行的DML操作都可以被复制，但是必须目标端目标表必须存在
添加：
GGSCI (sourceserver) > ADD EXTRACT PHREMD1, EXTTRAILSOURCE dirdat/l1
GGSCI (sourceserver) > ADD RMTTRAIL dirdat/l2, EXTRACT PHREMD1, MEGABYTES 100
#添加一个dp01,从哪里读取什么文件传输到哪里去
启动
start ext phremd1
二，目标端配置：
GGSCI>edit param mgr
PORT 7809
GGSCI (oracle11) 3> edit param ./GLOBALS
CHECKPOINTTABLE ogg.checkpoint ---ogg为管理用户
GGSCI>exit #这里要退出来,下面的ADD CHECKPOINTTABLE才能执行成功，否则会报CHECKPOINTTABLE不存在
GGSCI>DBLOGIN USERID ogg, PASSWORD ogg ## ADD replicat checkpoint table in target system
GGSCI>add checkpointtable ogg.checkpoint
GGSCI (10gasm) 3> view param rhremd1 #编辑配置文件，告诉他做什么
Replicat RHREMD1
-------------------------------------------------------------------
-- Replicat for HR Schema
-------------------------------------------------------------------
SETENV (NLS_LANG = AMERICAN_AMERICA.AL32UTF8)
USERID ogg, PASSWORD ogg
HandleCollisions
AssumeTargetDefs
Map HR.*, Target HR.* ; ---注意：空格和分号---结构必须相同，名字可以不同
GGSCI>ADD REPLICAT RHREMD1, EXTTRAIL dirdat/l2 #要指明提取的队列文件(源端会把数据发送到这里)
启动
start rep rhremd1
三，测试：
对源端hr.employees 表进行增删改，查看目标端的数据是否发生变化。
四，总结
两点注意：使用ASM管理的数据库在源端配置抽取进程的时候必须指定TRANLOGOPTIONS
配置文件中，MAP TEST.TEST, TARGET TEST.TEST; ---- 要注意TARGET前面空格和最后的分号不能少
关于TRANDATA：http://www.itpub.net/thread-1332238-2-1.html
五，关于常用的维护命令
info all
info ext lhremd1
info ext lhremd1,detail
stats ext lhremd1
view report lhremd1
view param lhremd1
info mgr
edit param lhremd1