First 50 of 200 questions. Each one lists the choices, the correct answer, and an explanation. The checker at the top of this page saves progress in this browser when it is available.
Continue in the practice tool above to load more questions, 50 at a time.
Question 1Database Architecture and Configuration
A database instance is started with STARTUP MOUNT. Which component is already running at this stage, while user data files remain unopened?
- The user schemas' open data files
- The listener registration for remote clients
- The instance memory structures and background processes Correct answer.
- The database's online redo log threads
Explanation
The MOUNT stage starts the instance, which means the SGA is allocated and background processes are running, but only the control files are opened. Data files and redo log files are not opened until the OPEN stage. Listener registration is a service-level concern, not an instance start stage.
Question 2Database Architecture and Configuration
Which memory area stores the parsed representation of SQL statements and the execution plans for reuse by multiple sessions?
- The PGA
- The Java pool
- The shared pool Correct answer.
- The large pool
Explanation
The shared pool holds the library cache, which stores parsed SQL, PL/SQL, and execution plans, plus the data dictionary cache. Because it lives in the SGA, cursors there can be shared across sessions. The PGA is private per-session memory and cannot be shared.
Question 3Database Architecture and Configuration
You want the buffer cache to use automatic memory management based on a target size. Which initialization parameter must be set for this behavior?
- DB_CACHE_SIZE
- SGA_TARGET Correct answer.
- PGA_AGGREGATE_TARGET
- SHARED_POOL_SIZE
Explanation
Setting SGA_TARGET to a nonzero value enables automatic shared memory management, letting Oracle redistribute memory among the SGA components. Individual component parameters such as DB_CACHE_SIZE then act as lower bounds. PGA_AGGREGATE_TARGET governs program global area, not the SGA.
Question 4Database Architecture and Configuration
Which background process writes changed blocks from the buffer cache to data files during normal operation?
- CKPT
- DBWn Correct answer.
- LGWR
- ARCn
Explanation
The database writer process, DBWn, flushes dirty buffers from the buffer cache to the data files. LGWR writes redo to the online redo logs, CKPT updates checkpoint information, and ARCn archives redo log files. Only DBWn moves data blocks to data files.
Question 5Database Architecture and Configuration
A database runs in NOARCHIVELOG mode. What is the direct consequence for recovery?
- Online redo logs are never written in this mode
- Redo log switches are impossible until archiving is enabled
- The database can be backed up only while mounted
- Point-in-time recovery beyond the last full backup is not possible Correct answer.
Explanation
In NOARCHIVELOG mode, filled redo log files are overwritten rather than archived, so no redo history exists to roll forward past the last consistent backup. Recovery is limited to restoring that backup. Redo is still written and log switches still occur.
Question 6Database Architecture and Configuration
Which file type is required for the instance to start and contains the database name, control file locations, and other startup metadata?
- The alert log
- The password file
- A trace file
- The server parameter file or a text initialization parameter file Correct answer.
Explanation
The instance reads its initialization parameters from an SPFILE or PFILE at startup; these supply the database name, control file paths, and memory settings. Without valid parameter files the instance cannot be started. Password files authenticate privileged users, and logs record events.
Question 7Database Architecture and Configuration
Which view would you query to see the current value of an initialization parameter and whether it was changed in the session or system scope?
- V$PARAMETER Correct answer.
- V$DATABASE
- V$SESSION
- V$INSTANCE
Explanation
V$PARAMETER shows current parameter values, including whether a change applies to the session or the system. V$INSTANCE reports instance status, V$DATABASE reports database identity and mode, and V$SESSION reports sessions. Parameter inspection belongs to V$PARAMETER.
Question 8Database Architecture and Configuration
Which tablespace is created automatically when a database is created and stores the data dictionary base tables and views?
- SYSTEM Correct answer.
- USERS
- TEMP
- UNDO
Explanation
The SYSTEM tablespace is always created and contains the data dictionary and many core objects. SYSAUX is also created automatically as an auxiliary tablespace. TEMP, UNDO, and USERS are typically created during database creation but serve other purposes.
Question 9Database Architecture and Configuration
A database uses automatic undo management. What happens to undo data when a transaction commits?
- It is moved into the SYSTEM tablespace
- It is retained until the undo retention period expires Correct answer.
- It is deleted immediately to free space
- It is written to the alert log
Explanation
With automatic undo management, committed undo remains in the undo tablespace until it is no longer needed for read consistency and the retention period passes. It is then eligible for reuse. Undo is not deleted at commit and is never relocated to SYSTEM.
Question 10Database Architecture and Configuration
Which initialization parameter determines the maximum number of user processes that can connect to the instance?
- SESSIONS
- TRANSACTIONS
- PROCESSES Correct answer.
- OPEN_CURSORS
Explanation
PROCESSES limits operating system processes, including background and user processes, that can attach to the instance. SESSIONS is derived from PROCESSES and limits sessions, while TRANSACTIONS and OPEN_CURSORS govern other per-session resources. Exceeding PROCESSES prevents new connections.
Question 11Database Architecture and Configuration
An administrator runs ALTER SYSTEM SET DB_CACHE_SIZE=512M SCOPE=MEMORY. What is the effect?
- The change is discarded at the next log switch
- The change is written to the SPFILE and survives restart
- The change applies to the current instance only and is lost at restart Correct answer.
- The change applies only to the current session
Explanation
SCOPE=MEMORY changes the parameter in the running instance without updating the SPFILE, so the new value is lost when the instance restarts. SCOPE=SPFILE would persist the change, and SCOPE=BOTH would do both. DB_CACHE_SIZE is not a session-modifiable parameter.
Question 12Database Architecture and Configuration
You want the instance to start automatically and be opened when the server boots on a Linux system using Oracle Restart. Which action is required?
- Set the parameter REMOTE_LOGIN_PASSWORDFILE to EXCLUSIVE
- Add STARTUP to the alert log
- Create a database trigger that runs STARTUP
- Register the database with SRVCTL and enable it Correct answer.
Explanation
Oracle Restart uses the SRVCTL utility to manage resources; you add the database with srvctl add database and enable it so it starts at boot. The alert log is a log, not a startup mechanism, and triggers cannot start the instance. Password file settings relate to authentication.
Question 13Database Architecture and Configuration
Which statement about the SYSAUX tablespace is correct?
- It replaces the SYSTEM tablespace for the data dictionary
- It is mandatory and must be created before the database can open
- It is created automatically at database creation and hosts auxiliary components such as AWR and Statspack Correct answer.
- It stores only user-created tables
Explanation
SYSAUX is created automatically as an auxiliary tablespace for components such as Automatic Workload Repository and Statspack, reducing load on SYSTEM. It is required in current releases but is created by the database creation process. The core data dictionary remains in SYSTEM.
Question 14Database Architecture and Configuration
A tablespace is created with SEGMENT SPACE MANAGEMENT AUTO. What does this setting control?
- Extents are allocated in fixed sizes regardless of the segment
- Blocks are compressed automatically on insert
- Undo retention is calculated automatically
- Free space within segments is tracked with bitmaps instead of free lists Correct answer.
Explanation
Automatic segment space management uses bitmaps in segment headers to track free space, replacing free lists and improving concurrency. It does not control extent sizing or compression. Undo retention is governed by undo parameters, not segment space management.
Question 15Database Architecture and Configuration
You want to create a tablespace whose data files grow automatically up to a defined maximum. Which clause should you use?
- AUTOEXTEND ON NEXT 100M MAXSIZE 2G Correct answer.
- SEGMENT SPACE MANAGEMENT AUTO
- EXTENT MANAGEMENT LOCAL UNIFORM SIZE 1M
- DEFAULT STORAGE (INITIAL 10M)
Explanation
The AUTOEXTEND ON clause, together with NEXT and MAXSIZE, allows a data file to grow automatically up to the specified maximum. Extent management controls allocation within the tablespace, not file growth. Storage defaults apply to segments, not data files.
Question 16Database Architecture and Configuration
A database is running in ARCHIVELOG mode. Which command sequence places it in that mode from a clean shutdown?
- ALTER DATABASE ARCHIVELOG; SHUTDOWN IMMEDIATE; STARTUP
- STARTUP; ALTER DATABASE ARCHIVELOG; ALTER SYSTEM SWITCH LOGFILE
- STARTUP NOMOUNT; ALTER DATABASE ARCHIVELOG; ALTER DATABASE OPEN
- STARTUP MOUNT; ALTER DATABASE ARCHIVELOG; ALTER DATABASE OPEN Correct answer.
Explanation
Changing the archiving mode requires the database to be mounted but not open. You start in MOUNT, issue ALTER DATABASE ARCHIVELOG, then open the database. The mode change cannot be made while the database is fully open or in NOMOUNT.
Question 17Database Architecture and Configuration
A user reports that a query needs a large sort area and the session is spilling to disk. Which parameter most directly influences the size of a dedicated server session's work area?
- PGA_AGGREGATE_TARGET Correct answer.
- DB_FILE_MULTIBLOCK_READ_COUNT
- SGA_TARGET
- CURSOR_SHARING
Explanation
PGA_AGGREGATE_TARGET sets the total PGA memory target, from which Oracle derives work area sizes such as sort and hash areas for dedicated sessions. SGA_TARGET sizes the shared memory, not private work areas. Multiblock read count affects I/O, and cursor sharing affects parsing.
Question 18Database Architecture and Configuration
Which view reports the current open mode and role of the database, such as READ WRITE or MOUNTED?
- V$TABLESPACE
- V$DATABASE Correct answer.
- V$LOG
- V$INSTANCE
Explanation
V$DATABASE includes columns such as OPEN_MODE and DATABASE_ROLE, which describe whether the database is mounted, read write, or read only. V$INSTANCE reports instance status, V$TABLESPACE lists tablespaces, and V$LOG describes redo log groups. Open mode is a database-level attribute.
Question 19Database Architecture and Configuration
You create a tablespace with EXTENT MANAGEMENT LOCAL and UNIFORM SIZE 1M. What is the effect on extent allocation?
- Each segment receives a 1M initial extent and then sizes are chosen by Oracle
- Extents are allocated in 1M units for every segment in the tablespace Correct answer.
- Segment free space is tracked in 1M bitmaps
- The tablespace's data files are limited to 1M each
Explanation
With local extent management and a uniform size, every extent allocated in the tablespace is exactly that size, regardless of the segment. This is common for undo and temporary tablespaces. It does not limit data file size or affect space management bitmaps.
Question 20Database Architecture and Configuration
Which statement about the SPFILE is correct?
- It is a text file that administrators edit directly with an editor
- It stores parameter values persistently and is updated by ALTER SYSTEM Correct answer.
- It is created only when the database is in ARCHIVELOG mode
- It can be modified only when the instance is shut down
Explanation
The SPFILE is a binary server-side parameter file maintained by the server; ALTER SYSTEM with SCOPE=SPFILE or SCOPE=BOTH updates it. Administrators do not edit it directly. It can be updated while the instance is running and is unrelated to archiving mode.
Question 21Database Architecture and Configuration
A database uses a small undo tablespace and long-running queries fail with ORA-01555. Which configuration change most directly addresses the root cause?
- Increase SHARED_POOL_SIZE
- Enable FORCE LOGGING on the database
- Reduce the size of the redo log files
- Increase the size of the undo tablespace and set an appropriate UNDO_RETENTION Correct answer.
Explanation
ORA-01555 occurs when required undo has been overwritten before a query finishes. Providing more undo space and a retention period long enough for the longest queries reduces the chance of overwrite. Shared pool, force logging, and redo log sizing do not preserve undo records.
Question 22Database Architecture and Configuration
Which component of the SGA holds information about recently executed SQL and PL/SQL so that repeated executions can avoid re-parsing?
- The result cache
- The data dictionary cache
- The library cache Correct answer.
- The redo log buffer
Explanation
The library cache, part of the shared pool, stores parsed SQL and PL/SQL and their execution plans for reuse. The data dictionary cache stores object metadata, the redo log buffer stages redo, and the result cache stores query results. Parsing reuse is a library cache function.
Question 23Database Architecture and Configuration
You must reduce the size of the buffer cache in an instance using automatic shared memory management. Which approach is valid?
- Restart the instance in NOMOUNT mode
- Set DB_CACHE_SIZE to a lower value and shrink the SGA_TARGET accordingly Correct answer.
- Issue ALTER SYSTEM FLUSH BUFFER_CACHE
- Drop and recreate the SYSTEM tablespace
Explanation
Under automatic shared memory management, component sizes are adjusted through their parameters, and lowering SGA_TARGET lets Oracle redistribute memory. DB_CACHE_SIZE can be set as a lower bound. Flushing the buffer cache empties it without resizing, and restarting in NOMOUNT does not change the target.
Question 24Database Architecture and Configuration
A data file has AUTOEXTEND ON but the underlying file system is full, and inserts fail with an unable-to-extend error. Which action resolves the immediate problem?
- Change the tablespace to SEGMENT SPACE MANAGEMENT MANUAL
- Set the parameter DB_CACHE_SIZE to a larger value
- Increase the value of DB_BLOCK_SIZE
- Add space to the file system or add a new data file to the tablespace Correct answer.
Explanation
AUTOEXTEND can only grow a file if the operating system has free space. Freeing space on the file system or adding a data file to the tablespace provides the needed capacity. Block size, segment space management, and buffer cache size do not create storage.
Question 25Database Architecture and Configuration
An instance is started with STARTUP RESTRICT. Which statement describes the resulting state?
- All users can connect but cannot perform DML
- Only users with the RESTRICTED SESSION privilege can connect Correct answer.
- Only SYSDBA can connect, and only in NOMOUNT mode
- The database is mounted but never opened
Explanation
STARTUP RESTRICT opens the database but allows connections only to users granted the RESTRICTED SESSION privilege, such as DBAs. It is useful for maintenance that requires exclusive access. Other users can connect after the restriction is lifted with ALTER SYSTEM DISABLE RESTRICTED SESSION.
Question 26Database Architecture and Configuration
An instance is configured with MEMORY_TARGET set to 2G and no explicit SGA_TARGET or PGA_AGGREGATE_TARGET values. Which statement describes how memory is managed?
- The SGA is fixed at 2G and the PGA is allocated separately from operating system memory outside the target.
- Oracle automatically tunes the total SGA and instance PGA sizes within the 2G target, redistributing memory between them as workload demands change. Correct answer.
- The parameter is ignored unless SGA_TARGET is also set, so the instance uses default component sizes.
- Only the buffer cache is auto-tuned; the shared pool and PGA remain at their default fixed sizes.
Explanation
MEMORY_TARGET enables automatic memory management, in which Oracle dynamically adjusts the aggregate SGA and aggregate PGA within the single target. Internally it sets SGA_TARGET and PGA_AGGREGATE_TARGET, and memory can shift between the two areas as demand changes.
Question 27Database Architecture and Configuration
A database is created with a server parameter file. An administrator wants a changed parameter value to persist across instance restarts but does not want it applied to the currently running instance. Which SCOPE value should be used with ALTER SYSTEM SET?
- SCOPE=BOTH
- SCOPE=MEMORY
- SCOPE=DEFERRED
- SCOPE=SPFILE Correct answer.
Explanation
SCOPE=SPFILE writes the change only to the server parameter file, so it takes effect at the next startup and leaves the running instance untouched. SCOPE=BOTH would apply it immediately as well as persist it, which is not what is requested.
Question 28Database Architecture and Configuration
You are planning storage for a new tablespace and want Oracle to manage free space within each segment using bitmaps rather than free lists. Which tablespace attribute must be specified?
- UNIFORM SIZE 1M
- SEGMENT SPACE MANAGEMENT AUTO Correct answer.
- EXTENT MANAGEMENT DICTIONARY
- SEGMENT SPACE MANAGEMENT MANUAL
Explanation
SEGMENT SPACE MANAGEMENT AUTO uses bitmaps stored in the segment header to track free and used space in blocks, replacing free lists. This is the default for locally managed tablespaces and generally reduces contention on segment headers.
Question 29Database Architecture and Configuration
A production database must be protected so that a media failure can be recovered to the point of failure. Which configuration is required?
- The database runs in NOARCHIVELOG mode with frequent cold backups.
- The database uses a SPFILE and a control file multiplexed across three locations.
- The database has FAST_START_MTTR_TARGET set to a low value.
- The database runs in ARCHIVELOG mode with the archived redo logs retained on durable storage. Correct answer.
Explanation
Point-in-time recovery to the point of failure requires the archived redo logs that contain all changes since the last backup, so the database must be in ARCHIVELOG mode and those logs must be preserved. Control file multiplexing and MTTR targets improve availability but do not supply the missing redo.
Question 30Database Architecture and Configuration
A tablespace uses automatic segment space management and contains segments with very different row sizes. Which statement about its blocks is correct?
- Blocks are always filled completely before a new block is used, regardless of PCTFREE.
- Free space is tracked with free lists whose count is set by FREELISTS at segment creation.
- Free space is tracked with bitmaps, and PCTUSED is ignored while PCTFREE still reserves space for row growth. Correct answer.
- The PCTFREE and PCTUSED values together determine when a block is returned to the free list.
Explanation
With automatic segment space management, free space is tracked by bitmaps in the segment header and PCTUSED has no meaning. PCTFREE is still honored to reserve room in a block for updates that enlarge existing rows.
Question 31Database Architecture and Configuration
An administrator creates a bigfile tablespace with a single 500G data file. Which statement about this tablespace is correct?
- It can contain only one data file, which may be very large, and it must be locally managed with automatic segment space management. Correct answer.
- It must use dictionary extent management because of its size.
- It requires the data file to be placed on a raw device.
- It can contain up to 1022 data files, each limited to 4M blocks.
Explanation
A bigfile tablespace has exactly one data file that can be extremely large, and it is always locally managed with automatic segment space management. This simplifies management by reducing the number of files to track.
Question 32Database Architecture and Configuration
A database has three online redo log groups with one member each, and a disk failure destroys the current redo log group. Which statement describes the consequence?
- The instance continues running and automatically recreates the lost member from the control file.
- The database remains open but all transactions in the lost group are silently rolled back.
- The instance may terminate, and recovery requires clearing the lost log group with ALTER DATABASE CLEAR LOGFILE. Correct answer.
- The instance switches to the next group and the lost group is dropped automatically at the next checkpoint.
Explanation
A single-member redo log group is a single point of failure; losing the current group can crash the instance, and the group must be cleared before the database can be opened normally. Multiplexing members with LGWR to separate disks prevents this situation.
Question 33Database Architecture and Configuration
Which background process is responsible for writing redo log buffer entries to the online redo log files?
- DBWn
- CKPT
- LGWR Correct answer.
- ARCn
Explanation
LGWR writes the redo log buffer to the online redo log files, flushing on commit, when the buffer is one-third full, and at other triggering events. This makes committed changes recoverable even before the corresponding data blocks are written.
Question 34Database Architecture and Configuration
A tablespace is created with EXTENT MANAGEMENT LOCAL AUTOALLOCATE. Which statement describes extent sizing?
- Extent sizes are chosen by Oracle, starting small and increasing in steps as the segment grows. Correct answer.
- Extent sizes are determined by the NEXT storage clause of each segment.
- Every extent is exactly 64K regardless of segment growth.
- Extent sizes are recorded in the data dictionary and can be changed with ALTER TABLESPACE.
Explanation
With AUTOALLOCATE, Oracle selects extent sizes, beginning with small extents and increasing them as the segment grows, which reduces wasted space for small objects. It cannot be combined with a uniform extent size in the same tablespace.
Question 35Database Architecture and Configuration
A database must be opened read-only so that users can query data while redo from the primary is being applied. Which statement about this state is correct?
- The database must be in NOARCHIVELOG mode before it can be opened read-only.
- Data files are opened read-only, queries are allowed, and recovery can continue applying redo in the background. Correct answer.
- The database is mounted but data files remain closed, so no queries are possible.
- The database is fully open for writes but only one session may modify data at a time.
Explanation
Opening a physical standby database read-only allows queries against the data files while managed recovery continues applying redo. This supports reporting offload without interrupting redo apply.
Question 36Database Architecture and Configuration
You want a tablespace's data file to be created with a specific initial size and to extend automatically in fixed increments. Which pair of clauses accomplishes this?
- DATAFILE 100M and EXTENT MANAGEMENT LOCAL
- SIZE 100M and AUTOEXTEND ON NEXT 50M Correct answer.
- INITIAL 100M and MAXSIZE 50M
- SIZE 100M and RESIZE 50M
Explanation
The SIZE clause sets the initial data file size, and AUTOEXTEND ON NEXT defines the increment added each time the file must grow. MAXSIZE optionally caps growth, but it does not set the increment.
Question 37Backup and Recovery
In an Oracle multitenant database, which tool must you use to back up and recover a pluggable database while it is open and being used?
- RMAN connected to the root as a common user Correct answer.
- Operating system file copy of the PDB datafiles
- SQL*Plus startup mount of the PDB
- Oracle Data Pump export of the PDB schema
Explanation
RMAN is the supported mechanism for backing up an open PDB, and it must be invoked through a connection to the root container with a common user such as SYS or a user granted SYSDBA or SYSBACKUP. A PDB cannot be placed in mount state independently, so file-copy and SQL*Plus approaches are not valid for an open PDB.
Question 38Backup and Recovery
Which command creates an image copy of the database that RMAN can subsequently use for incremental updates?
- BACKUP AS COPY DATABASE Correct answer.
- BACKUP DATAFILE ALL
- BACKUP INCREMENTAL LEVEL 0 DATABASE
- BACKUP ARCHIVELOG ALL
Explanation
BACKUP AS COPY DATABASE produces image copies, which are byte-for-byte duplicates of the datafiles and can later be updated with an incremental backup applied to the copy. A level 0 incremental backup creates a backup set, not an image copy, so it cannot be rolled forward the same way.
Question 39Backup and Recovery
You want RMAN to automatically delete archived redo logs from the fast recovery area after they have been backed up at least once and are no longer needed. Which configuration accomplishes this?
- CONFIGURE ARCHIVELOG DELETION POLICY TO APPLIED ON STANDBY
- CONFIGURE RETENTION POLICY TO REDUNDANCY 1
- CONFIGURE BACKUP OPTIMIZATION ON
- CONFIGURE ARCHIVELOG DELETION POLICY TO BACKED UP 1 TIMES TO DISK Correct answer.
Explanation
The archived log deletion policy TO BACKED UP n TIMES TO device tells RMAN that archived logs become eligible for deletion once they have been backed up the specified number of times to that device. This policy is evaluated when the fast recovery area needs space or during DELETE operations.
Question 40Backup and Recovery
A datafile is accidentally deleted and the database is still open. Which RMAN command restores the missing datafile to its original location without applying any redo?
- RECOVER DATAFILE 7
- RESTORE DATABASE
- SWITCH DATAFILE 7
- RESTORE DATAFILE 7 Correct answer.
Explanation
RESTORE DATAFILE recreates the datafile from a backup at its configured location but does not apply redo, so the file is left at the backup SCN. Recovery must follow to bring the file forward to a consistent point before the tablespace can be brought online.
Question 41Backup and Recovery
Which file must exist and be accessible for RMAN to perform an automatic backup of the control file?
- A standby control file on the target host
- A snapshot control file location with sufficient space Correct answer.
- A password file for the target database
- A trace file in the diagnostic destination
Explanation
RMAN creates a snapshot control file whenever it needs a consistent read-consistent copy of the control file, for example during control file autobackup or resynchronization. If the snapshot location is unavailable or full, those operations fail.
Question 42Backup and Recovery
Which RMAN feature lets you recover a table or set of tables from a backup without restoring the entire database?
- Flashback Table
- Table recovery with RECOVER TABLE Correct answer.
- Block media recovery
- Tablespace point-in-time recovery
Explanation
The RECOVER TABLE command in RMAN restores the needed datafiles into a temporary auxiliary instance, extracts the table or tables, and imports them into the target, all without a full database restore. It can use a point-in-time or SCN target for the extraction.
Question 43Backup and Recovery
Which view should you query on the target database to determine whether RMAN backups exist and when they were taken, without connecting to a recovery catalog?
- V$ARCHIVED_LOG
- V$DATAFILE
- V$BACKUP
- V$RMAN_BACKUP_JOB_DETAILS Correct answer.
Explanation
V$RMAN_BACKUP_JOB_DETAILS in the target control file records completed RMAN backup jobs, including status and timing. It is available whenever the target is mounted or open and does not require a catalog connection.
Question 44Backup and Recovery
Which initialization parameter must be set to enable RMAN to use a fast recovery area for archived logs and backups?
- CONTROL_FILE_RECORD_KEEP_TIME
- DB_CREATE_FILE_DEST
- DB_RECOVERY_FILE_DEST Correct answer.
- LOG_ARCHIVE_DEST_1
Explanation
DB_RECOVERY_FILE_DEST specifies the fast recovery area location, and DB_RECOVERY_FILE_DEST_SIZE caps its size. Together these parameters let Oracle manage archived logs, backups, and flashback logs in that area.
Question 45Backup and Recovery
You issue BACKUP INCREMENTAL LEVEL 1 DATABASE after a level 0 backup. Which blocks does RMAN read to build this backup?
- All blocks in every datafile
- Only blocks recorded in the change tracking file
- Only blocks changed since the most recent level 0 or level 1 backup
- Only blocks changed since the most recent level 0 backup Correct answer.
Explanation
A level 1 incremental backup is differential by default, so RMAN backs up blocks changed since the most recent level 0 backup. A cumulative level 1 backup instead captures changes since the last level 0 or level 1 backup.
Question 46Backup and Recovery
You want RMAN to skip reading datafile blocks that have not changed since the last backup, to reduce backup time. Which configuration enables this?
- CONFIGURE CONTROLFILE AUTOBACKUP ON
- CONFIGURE DEVICE TYPE DISK PARALLELISM 2
- CONFIGURE BACKUP OPTIMIZATION ON
- CONFIGURE BLOCK CHANGE TRACKING USING FILE Correct answer.
Explanation
Enabling block change tracking creates a change tracking file that records modified blocks, allowing RMAN to read only those blocks during incremental backups. It does not change which blocks are logically included, only how RMAN finds them.
Question 47Backup and Recovery
A tablespace is lost and the database is open. Which sequence correctly recovers it with RMAN?
- RECOVER TABLESPACE then RESTORE TABLESPACE
- RESTORE TABLESPACE then ALTER TABLESPACE ONLINE
- RESTORE TABLESPACE then RECOVER TABLESPACE, then ALTER TABLESPACE ONLINE Correct answer.
- RESTORE DATABASE then OPEN RESETLOGS
Explanation
For tablespace recovery the tablespace must first be taken offline, then restored from backup, recovered to apply redo, and finally brought back online. Restoring without recovery leaves the files at an inconsistent SCN and the online step fails.
Question 48Backup and Recovery
You configure RMAN with two disk channels and issue BACKUP DATABASE. What does the parallelism setting primarily control?
- The number of concurrently active channels used for the backup Correct answer.
- The number of archived log destinations used
- The number of backup sets produced for each datafile
- The compression level applied to backup sets
Explanation
Channel parallelism determines how many server sessions perform the backup concurrently, which can improve throughput on suitable storage. It does not change the logical content of the backup.
Question 49Backup and Recovery
You need RMAN to retain enough backups to recover to any point in the last 14 days. Which retention configuration is most appropriate?
- CONFIGURE RETENTION POLICY TO REDUNDANCY 14
- CONFIGURE RETENTION POLICY TO RECOVERY WINDOW OF 14 DAYS Correct answer.
- CONFIGURE RETENTION POLICY TO NONE
- CONFIGURE ARCHIVELOG DELETION POLICY TO NONE
Explanation
A recovery window retention policy keeps the backups and archived logs required to satisfy point-in-time recovery to any SCN within the specified window. Redundancy, by contrast, keeps a fixed number of backup copies rather than a time-based recoverability guarantee.
Question 50Backup and Recovery
A data block is reported as corrupt by a query, and you have a recent backup. Which RMAN command repairs only that block while the database remains open?
- RECOVER DATAFILE 4
- RECOVER DATABASE
- BLOCKRECOVER DATAFILE 4 BLOCK 120 Correct answer.
- RESTORE DATAFILE 4
Explanation
BLOCKRECOVER restores and recovers a specific block or range of blocks from a backup while the rest of the datafile and database stay online. It is faster and less disruptive than restoring the entire datafile.