oracle表空间时间点恢复用户管理方式.docVIP

表空间时间点恢复 用户管理方式 实验准备 1、 准备两个表空间及表 Create tablespace user02 Datafile 7../user02.dbf.’ size 10m Create tablespace user04 Datafile 7../user04.dbf’ size 10m Create table scott.customers(tablespace user02) Create table scott.sales(tablespace user04) 2、 备份主数据库的数据文件和控制文件到备份H录 Alter database begin backup Cp Alter database end backup 3、 备份控制文件 Alter databse backup control file to ‘/../controlOl.ctl’ Alter database mount clone database 4、 在主数据库上执行DML增加数据，多次切换曰志 Inser into scott.customers values() Commit Alter system switch logfile 5、记录当前恢复时间及日志序号 Select to_char(sysdate，’yyyy-mm-dd hh24:mi:ss’)from dual; Select SEQUENCE#,MEMBERS,ARCHIVED,STATUS from v$log; 6、用户误操作 Truncate table scott.cutomers; Inser into scott.sales values() Commit; Create tablescott.empl() 检査TSPITR之后将会丢失的数据 Select owner,name from ts_pitr_objects_to_be_dropped where tablespace_name=’ US ER02 ’ and creation_time〉to_date(62012-06-05 09:36:47’，’yyyy-mm-dd hh24:mi:ss’) 建立辅助数据库口令文件 Cmd Orapwd file=/../ password=oracle entries=10 建立辅助例程参数文件 Sqlplus Conn / as sysdba Create pfile=’/../initauxi.ora’ from spfile Shutdown immediate 编辑参数文件initaux.ora orcl. db cache size=lorcl._java_pool_size=4194304 orcl._large一pool_size=4194304 orcl._shared_pool_size= 138412032 orcl._streams_pool_size=0 *.audit_file_dest=7u01/oracle/admin/orcl/adump *.background_dump_dest=’/u01/oracle/admin/orcl/bdump .compatible:’ .0’ .control_files=7u01 /oracle/auxi/controlO l.ctl--更改 气 core_dump_dest=’/uO 1/oracle/admin/orcl/cdump’ *.db block size二8192 *.db domain=’’ *.db file multiblock read count: 16 *.db_name=’orcl’ 气db一recovery _file_dest=’/u01/oracle/flash一recovery _area’ 气 db_recovery_file_dest_size=2147483648 *.dispatchers=’(PROTOCOL=TCP) (SERVICE=orclXDB)’ *.j ob_queue_processes= 10 * .open_cursors=300 *.pga_aggregate_target气 processes=150 气remote_login_passwordfile=’EXCLUSIVE’ *.sga_target=262144000 气undo_management=’AUTO’ 气 undo_tablespace=UNDOTBS 1 *.user_dump_dest=/u01/oracle/admin/orcl/udump’ db_unique_name=’auxi’ --