User managed complete recovery | Oracle DBA Tutorial pdf

User managed complete recovery

In all the following cases, it is assumed that database is running in archive log mode and backup was already taken.
=> System table space lost/corrupt
1. STARTUP MOUNT
2. Restore only system table space file
3. If required, relocate file – ALTER DATABASE RENAME file TO new file
4. RECOVER AUTOMATIC DATABASE (or DATAFILE file name)
5. ALTER DATABASE OPEN
=>  Non-system table space lost/corrupt
1. STARTUP MOUNT
2. Make data file offline – ALTER DATABASE file name OFFLINE
3. ALTER DATABASE OPEN
4. Restore data file from backup to database location
5. If required, relocate file – ALTER TABLESPACE table space name
OFFLINE IMMEDIATE for issuing check point
6. RECOVER AUTOMATIC TABLESPACE table space name
7. ALTER TABLESPACE table space name ONLINE
=> Non-system table space lost/corrupt when no backup taken
1. STARTUP MOUNT
2. ALTER DATABASE CREATE DATAFILE filename with path AS new filename (only if relocate)
3. RECOVER DATABASE (DATAFILE filename)
4. ALTER DATABASE OPEN
=> Instance failed during backup
Entire database won’t start. Data file header freezes during backup, so no checkpoint information is written.
1. STARTUP MOUNT
2. ALTER DATABASE END BACKUP
3. ALTER DATABASE OPEN