RMAN异机迁移数据库.docxVIP

旧数据库备份文件查看force_logging信息，该参数在数据库全备前设为YESSQL select force_logging from v$database;如为NO，则修改为YESSQL alter database force logging;RMAN备份脚本run{allocate channel c1 type disk;allocate channel c2 type disk;backup database format /home/oracle/rman/full_%d_%T_%s_%p;sql alter system archive log current;backup archivelog all format /home/oracle/rman/arch_%d_%T_%s_%p delete input;backup current controlfile format /home/oracle/rman/controlfile_%d_%T_%s_%p;}全库备份$ rman target / nocatalog cmdfile=rman.sql log=rman.log $ tail -f rman.log根据RMAN日志记录归档序号时间，例如Starting?backup?at?20-JAN-2015?12:42:33Finished?backup?at?20-JAN-2015?12:44:18 查看归档序号，记录全备开始前时间点的最后一个序号，和全备完成时的第一个序号，如SQL select THREAD#, SEQUENCE# ,to_char(FIRST_TIME,yyyymmddhh24:mi),to_char(COMPLETION_TIME,yyyymmdd hh24:mi) from v$archived_log order by FIRST_TIME;由结果可得，全备开始前最后一个归档序号为13，全备完成的第一个序号为14，记录线程1的归档序号范围为13-14，如存在多个线程，记录方法类似创建原数据库的PFILE文件SQL create pfile=/oracle/pfilora from spfile;将备份集与参数文件复制到目标数据库的相同目录下$scp fullbak/* 192.168.6.22:/home/oracle/fullbak/$scp /oracle/pfilora 192.168.6.22:/oracle/新数据库恢复为了恢复数据库，在新数据库上拷贝得到的pfile参数文件需要添加*.local_listener=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=TCP)(HOST=ip)))，此处的ip就是oracle数据库所在服务器的ip否则会出现下图错误使用pfile文件启动数据库SQL startup nomount pfile=/oracle/pfilora;恢复控制文件RMAN restore controlfile from /home/oracle/fullbak/controlfile_UP15_1;MOUNT数据库SQL alter database mount;编写RMAN恢复脚本，如下oracle@dstdb:/home/oracle$more restore_database.sh run { allocate channel c1 device type disk; allocate channel c2 device type disk; allocate channel c3 device type disk; allocate channel c4 device type disk; set newname for datafile 1 to /oracle/app/oracle/oradata/up/system02.dbf;set newname for datafile 2 to /oracle/app/oracle/oradata/up/sysaux02.dbf;set newname for datafile 3 to /oracle/app/oracle/oradata/up/undotbs02.dbf;set newname for datafile 4 to /oracle/app/orac