Oracle数据库RMAN还原和恢复

Oracle数据库RMAN还原和恢复
disaster classification
   physical failure
       datafile,controfile,redo log
  logical mistake
       object
physical disaster situation
mount or open
   non-system tablespace
   undo tbs
nomount
  controlfile,all database file
mount
   system tablespace
   online redo log file
non-system tbs
  alter datafile offline
  alter tablespace offline
  restore&recover
  revert datafile online

crosscheck backupset;
list backupset summary;
validate backupset 860,953,1080,1081,1110;
validate backupset 860,953,1080,1081,1110 check logical;

non-system tbs 非系统表空间和数据文件恢复
rm -f users.dbf
alter database datafile 5 online;
run {
    allocate channel d1 device type disk;
    sql "alter database datafile 4 offline";
    restore datafile 4;
    recover datafile 4;
    sql "alter database datafile 4 online;
}

undo表空间恢复与还原　必要让数据库知道undo不能用了
alter datafile 2 online
alter datafile 2 offline
rman中undo表空间不能offline
run {
　allocate channel d1 device type disk;
  restore datafile 2;
  recover datafile 2;
  }
在sqlplus中
select status from v$datafile
recover datafile 2;
alter database datafile 2 online;

system tbs  系统表空间恢复
startup mount
restore datafile 1;
recover datafile 1;
alter database open;
rm -f system01.dbf
startup mount force;
run {
  allocate channel d1 device type disk;
  restore datafile 1;
  recover datafile 1;
  sql "alter database open";
}
select open_mode from v$database;

controlfile  控制文件恢复
startup target with nomount
connect to catalog and target
restore controlfile,recover database
alter database open resetlogs or noresetlogs

rm -f control*
startup nomount force;
重新连接的时候有时候不能指目标数据库
还原控制文件只能resetlogs;
rman catalog rmanuser/qkhhfqp@pretty target sys/qkhhfqp
run {
 allocate channel d1 devie type disk;
 restore controlfile;
 sql "alter database mount";
 restore database;
 recover database ;
 sql "alter database open resetlogs";
}
run {
 allocate channel d1 devie type disk;
 restore database;
 sql "alter database mount";
 recover database ;
 sql "alter database open resetlogs";
}

online redo logfile   在线日志恢复
startup mount force
connect to target and catalog
list backupset of archivelog all
restore database
  recover database until sequence
10g之前resetlog之前备份将无效

rm -f redo*
startup mount force
run {
  allocate channel d1 device type disk;
  restore database;
  recover database;
}
alter database open resetlogs;
recover database until logseq=32
alter database open resetlogs;

all database   全部文件文件掉失恢复
nomount
   restore database
mount
   recover database
recover database until logseq=32(logseq的值)
alter database open resetlogs;

注意删除过期的备份
rm -f *
startup nomount force;
run {
  allocate channel d1 device type disk;
  restore database;
  sql "alter database mount";
  recover database;
}
change backupset 121,122 delete;
change archivelog all delete;

来自 “ ITPUB博客 ” ，链接：http://blog.itpub.net/11364208/viewspace-310414/，如需转载，请注明出处，否则将追究法律责任。

转载于:http://blog.itpub.net/11364208/viewspace-310414/