RMAN不完全恢复实战-优快云博客

本文详细解析了一次使用RMAN进行不完全恢复的真实案例。情境设定为2007年11月15日上午，因误操作导致USER_DATA表空间丢失，在确认重要交易后发生此事故。通过精确的时间点恢复策略，文章展示了如何恢复至11:00AM的数据库状态，包括控制文件的恢复、数据库的还原与恢复过程。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

You are using Recovery Manager (RMAN) for backup and recovery operations with a recovery
catalog. You have been taken database backups every evening. On November 15, 2007, at 11:30 AM,you were informed that the USER_DATA tablespace was accidentally dropped.
On investigation, you found that the tablespace existed until 11:00 AM, and important transactions were done after that.
So you decided to perform incomplete recovery until 11:00 AM. All the archive logs needed to perform recovery are intact. In NOMOUNT state you restored the control file that has information about the USER_DATA tablespace from the latest backup. Then you mounted the database.
Identify the next set of commands that are required to accomplish the task?
A.RMAN> run
{
SET UNTIL TIME 'Nov 15 2007 11:00:00';
RESTORE DATABASE;
RECOVER DATABASE;
}
B.RMAN> run
{
SET UNTIL TIME 'Nov 15 2007 11:00:00';
RESTORE DATABASE;
RECOVER DATABASE USING BACKUP CONTROLFILE;
}
C.RMAN> run
{
RESTORE DATABASE;
RECOVER DATABASE UNTIL TIME 'Nov 15 2007 11:00:00';
}
D.RMAN> run
{
RESTORE TABLESPACE user_data;
RECOVER TABLESPACE user_data UNTIL TIME 'Nov 15 2007 11:00:00';
}

　　此题考点是不完全恢复。由于进行了大量的事务，显然是要进行数据库级别的不完全的恢复，将所有相关的事务均回滚。
　　在数据库 mount 之后，指定时间点或 SCN 或 log 号，就可以进行恢复了。 A 正确。 BCD 的语法错误恢复完成后以 resetlogs 方式打开数据库