• 使用数据泵+dblink迁移数据库,适用于本地空间不足的情况


    col name for a40
    select name,locks,pins
    from v$db_object_cache
    where locks > 0 and pins > 0 and type='PROCEDURE';
    
    
    
    
    
    10.21.1.20 和10.24.51.11两个数据库的dblibk已经建好了,主要就impdp就可以了,一共需要导出
    
    
    
    drop user ILEARN_TRA       cascade;
    drop user ILA_SAP          cascade;
    drop user ILEARN_JOB       cascade;
    drop user ILEARN           cascade;
    drop user ILEARN_BEIYAN    cascade;
    drop user APEX_PUBLIC_USER cascade;
    
    
    srvctl stop database -d ilndb
    srvctl start database -d ilndb
    
    
    SELECT SID,JOB FROM DBA_JOBS_RUNNING;
    MACHINE    
     PORT      
     TERMINAL  
     PROGRAM  
    
    select username,sid,SERIAL# from gv$session where username ='ILEARN_TRA';
    select username,sid,SERIAL# from v$session where username ='ILEARN_TRA';
    USERNAME                              SID    SERIAL#
    ------------------------------ ---------- ----------
    ILEARN_TRA                           2545          5
    ALTER SYSTEM KILL SESSION '2545,5';
    SELECT SID,SERIAL# FROM V$SESSION WHERE USERNAME='ILEARN_TRA';
    
    
    这几个用户下数据,正式迁移的时候,10.21.1.20上这几个用户数据要删除,整个过程耗时10:15:38 
    
    这是上次测试的日志 
    
    
    
    SYS@ilndb1>  select name from v$database@dblink_datapump; 
    
    NAME 
    --------- 
    ILNDB 
    
    dblink导库过程: 
    
    1、在客户端数据库创建 dblink 和要保存数据文件的目录: 
    create public database link dblink_datapump connect to system identified by Clic1234 using '10.24.51.12:1521/ilndb'; 
    select name from v$database@dblink_datapump; 
    create or replace directory expdp_dir as '/oracle/expdp'; 
    grant read, write on directory expdp_dir to public; 
    
    2、在客户端数据库创建相关表空间和用户 
    create tablespace ILEARN_DATA DATAFILE '+DATADG' SIZE 32767M AUTOEXTEND off;        //对比生产端,加足表空间容量 
    create user ILEARN_TRA identified by "Qwer4321" default tablespace ILEARN_DATA; 
    grant resource,connect to ILEARN_TRA; 
    
    3、在客户端数据库使用IMPDP导入 
    nohup impdp system/Clic1234 network_link=dblink_datapump logfile=impdp_ilearn1.log directory=expdp_dir schemas=ILEARN_TRA EXCLUDE=statistics cluster=N job_name=SYS_IMPORT_ILEARN_TRA parallel=20 &
    select sum(bytes/1024/1024) from dba_segments where owner='ILEARN_TRA';
    impdp system/Clic1234 network_link=dblink_datapump logfile=impdp_ilearn3.log directory=expdp_dir schemas=ILEARN EXCLUDE=statistics cluster=N job_name=SYS_IMPORT_ILEARN  parallel=20
    select sum(bytes/1024/1024) from dba_segments where owner='ILEARN';
    
    impdp system/Clic1234 network_link=dblink_datapump logfile=impdp_ilearn2.log directory=expdp_dir schemas=ILA_SAP EXCLUDE=statistics cluster=N job_name=SYS_IMPORT_ILA_SAP parallel=14
    --wancheng
    
    
    impdp system/Clic1234 network_link=dblink_datapump logfile=impdp_ilearn4.log directory=expdp_dir schemas=ilearn_beiyan EXCLUDE=statistics cluster=N job_name=SYS_IMPORT_ilearn_beiyan parallel=14
    --wancheng 
    
    impdp system/Clic1234 network_link=dblink_datapump logfile=impdp_ilearn4.log directory=expdp_dir schemas=ilearn_job EXCLUDE=statistics cluster=N job_name=SYS_IMPORT_ilearn_job parallel=14
    --wancheng
    
    impdp system/Clic1234 network_link=dblink_datapump logfile=impdp_ilearn5.log directory=expdp_dir schemas=APEX_PUBLIC_USER EXCLUDE=statistics cluster=N job_name=SYS_IMPORT_ilearn_apex parallel=14
    wancheng
    
    
    4、在客户端数据库删除 dblink: 
    drop public database link dblink_datapump; 
      祝工作顺利,万事如意~~ 
    
      
      
      
    nohup impdp system/Clic1234 network_link=dblink_datapump logfile=impdp_ilearn_1210.log directory=expdp_dir schemas=ILEARN_TRA,ILA_SAP,ILEARN_JOB,ILEARN,ILEARN_BEIYAN,APEX_PUBLIC_USER  EXCLUDE=statistics cluster=N job_name=SYS_IMPORT_1210 parallel=14 &
    
    ILEARN_TRA,ILA_SAP,ILEARN_JOB,ILEARN,ILEARN_BEIYAN,APEX_PUBLIC_USER
    
    
    create user VIEW_LMS_WORKBOOKS identified by Clic1234 default tablespace ILEARN_DATA account unlock;
    CREATE_AND_VIEW_LMS_WORKBOOKS
  • 相关阅读:
    Python批量 png转ico
    线性回归
    【论文集合】多语言机器翻译
    【论文集合】curriculum learning总结/课程学习论文
    crf++分词
    强化学习的探索策略方式
    关于multiprocessing中的logging的数据打印到本地失败或重复问题
    置信区间绘图、以10次平均值为例
    打印CSDN博客内容JS
    SUMO学习笔记(3)
  • 原文地址:https://www.cnblogs.com/wangrongxin/p/6209740.html
Copyright © 2020-2023  润新知