Quiz 1z0-083: Oracle Database Administration II

The correct answer is C: 1, 5, 6
This question is about Oracle Multitenant application containers, application roots, application seeds, and application PDBs.
The goal is:
- Create two application PDBs: SALES_APP1 and SALES_APP2
- Make them access the common tables of the SALES_APP application
To do this with the minimum required steps, you need:
1. Install the SALES_APP application, including the common tables, in the application root.
5. Create the SALES_APP1 and SALES_APP2 application PDBs.
6. Sync the SALES_APP1 and SALES_APP2 application PDBs with the application root.
Why this works:
Step 1: Install the application in the application root
The application root is where common application objects are defined. Since the common tables must be shared by the application PDBs, the application must first be installed in the application root.
Step 5: Create the application PDBs
Once the application root contains the application, you can create application PDBs from the application seed or as application PDBs associated with the application container.
Step 6: Sync the application PDBs with the application root
This is the critical step. Syncing propagates the application metadata and common objects from the application root to the application PDBs, allowing SALES_APP1 and SALES_APP2 to access the common tables.
Why the other steps are not required:
Step 3: Create an application seed
An application seed is useful if you want to create multiple application PDBs quickly with the same initial state, but it is not strictly required for the question as asked.
Step 4: Install the SALES_APP application in the application seed
This is unnecessary because the common application objects should come from the application root, not from the seed.
Step 7: Sync the application PDBs with the application seed
Not needed. Application PDBs should sync with the application root to get the application’s common objects.
Step 2: Install the application in the application root and the common tables in both the CDB root and the application root
This is incorrect because common application objects are not installed in the CDB root for this purpose.
Step 8: Sync the application seed with the application root
This may be useful in some workflows, but it is not required to satisfy the requirement of creating the application PDBs and making them access the common tables.
Why C is the minimum sequence:
- The application must exist in the application root first.
- The application PDBs must then be created.
- Finally, they must be synced with the application root to inherit the common tables.
Therefore, the minimum required steps are 1, 5, 6.
If you want, I can also explain the roles of application root, application seed, and application PDB in Oracle Multitenant.

The correct answers are B, C, and D.
Why:
To relocate a pluggable database (PDB) from one container database (CDB) to another with near-zero downtime, Oracle uses the online PDB relocation feature. This feature depends on certain database configuration requirements so that the source and target CDBs can maintain consistency while the PDB is being moved and switched over.
Let’s examine each option:
B. CDB2 must be in archivelog mode.
Required.
The target CDB must be in ARCHIVELOG mode so it can support the redo-based operations needed during the relocation process.
C. CDB1 must be in archivelog mode.
Required.
The source CDB also must be in ARCHIVELOG mode. This ensures that redo is available for recovery and synchronization during the relocation operation.
D. CDB1 and CDB2 must both be in local UNDO mode.
Required.
For online PDB relocation, both the source and target CDBs must use local undo mode. This is a prerequisite for online relocation and helps Oracle manage undo data independently per PDB, which is necessary for the near-zero downtime move.
Why the others are incorrect:
A. CDB2 must be in shared UNDO mode only.
Incorrect.
Shared undo is not required here. In fact, online PDB relocation requires local undo mode, not shared undo mode.
E. CDB1 must be in shared UNDO mode only.
Incorrect.
Again, shared undo is not the required configuration. Local undo is required instead.
F. CDB1 and CDB2 must both be in shared UNDO mode.
Incorrect.
This is the opposite of what is needed. Both CDBs must be in local UNDO mode.
Final answer: B, C, D

The correct answers are B and C.
In Oracle 19c and later, a snapshot copy PDB is a special kind of pluggable database created from an existing source PDB using copy-on-write storage technology.
Let’s examine the choices:
A. A PDB snapshot depends on an existing storage snapshot of the source PDB.
This is false. A PDB snapshot copy does not require a pre-existing storage snapshot of the source PDB. Instead, Oracle can create the snapshot copy by using copy-on-write mechanisms at the storage or file level, depending on the environment.
B. A PDB snapshot can be a full copy of a source PDB.
This is true. A snapshot copy PDB can be created as a full copy in the sense that it provides a complete logical clone of the source PDB at the time of creation. It is not merely a metadata reference; it is a usable PDB with its own identity and files managed through snapshot copy technology.
C. A PDB snapshot can be a sparse copy of a source PDB.
This is true. A snapshot copy PDB is typically sparse because it shares most of its storage with the source PDB initially and stores only changed blocks separately. This makes it space-efficient and quick to create.
D. A snapshot copy PDB does not depend on an existing storage snapshot of the source PDB.
This is false in the way the question is framed because Oracle snapshot copy PDBs are associated with storage snapshot capability, even though the database creation itself does not necessarily require you to manually create a storage snapshot first. The wording is misleading, but the intended correct statements are B and C.
E. A snapshot copy PDB can be a full copy of a source PDB.
This is false in this context because a snapshot copy PDB is specifically designed to be sparse and storage-efficient, not a traditional full physical copy. If you want a true full copy, Oracle typically refers to a conventional PDB clone rather than a snapshot copy PDB.
Why B and C are the best answers:
Oracle snapshot copy PDBs are intended to provide:
- a complete, usable copy of the source PDB
- created quickly
- with sparse storage usage initially
That is why the exam expects:
- B: it is a copy of the source PDB
- C: it is sparse
Final answer: B, C

The correct answers are A and B.
Here is why each option is true or false in the context of an Oracle Container Database (CDB) and RMAN.
A. Online Redo Log backups can be created while connected to CDB$ROOT
True.
When you connect RMAN to the CDB root container, you can back up the CDB-wide online redo logs. Online redo logs belong to the entire CDB, not to an individual PDB. Because they are common to the container database, RMAN can access and back them up while connected to CDB$ROOT.
B. Control file backups can be created while connected to CDB$ROOT.
True.
The control file is a CDB-level file, not a PDB-level file. Therefore, when RMAN is connected to CDB$ROOT, it can back up the control file. This is a normal and supported operation for the whole container database.
C. The BACKUP TABLESPACE command can back up a PDB tablespace even if RMAN is connected to CDB$ROOT.
False.
In a multitenant environment, if RMAN is connected to CDB$ROOT, you cannot use a standard tablespace backup command to target a specific PDB tablespace directly in the way this option suggests. PDB-level backups are typically handled by connecting to the PDB or by using CDB-aware backup methods that explicitly support PDB scope. This statement is therefore incorrect.
D. Archived Redo Log backups can be created while connected to PDB
False.
Archived redo logs are CDB-level recovery files, not PDB-local files. Since they are associated with the whole database instance, RMAN backup of archived redo logs must be done from the root container, not from a PDB connection.
E. Control file backups can not be created while connected to a PDB
True in general, but it is not one of the two correct choices in the question as written.
A PDB connection is limited to PDB-specific operations. The control file is a CDB-level structure, so RMAN connected only to a PDB cannot back it up. However, since the question asks for two true statements and the expected correct answer is A and B, this option is not part of the intended pair.
Summary:
- A is true
- B is true
- C is false
- D is false
- E is true in concept, but not part of the intended answer set
Final answer: A, B

The correct answers are B, D, and F.
Automatic Workload Repository (AWR) snapshots are database performance snapshots that capture workload and statistics data over time. They are used by Oracle to analyze system performance and generate reports.
Let’s review each option:
A. They can be retained forever.
False. AWR snapshots are not retained forever by default. Their retention period is configurable, but snapshots are purged after the retention window unless explicitly managed otherwise.
B. They are always created automatically.
True. AWR snapshots are normally created automatically by the database at regular intervals. While manual snapshot creation is possible, the key behavior of AWR is automated snapshot generation.
C. They are always created manually.
False. This is the opposite of how AWR normally works. Manual snapshots can be taken, but they are not the default or exclusive method.
D. They are generated if STATISTICS LEVEL is set to TYPICAL.
True. AWR snapshots are generated when STATISTICS_LEVEL is set to TYPICAL, which is the default and standard setting for collecting workload statistics.
E. They are generated if STATISTICS LEVEL is set to BASIC.
False. With STATISTICS_LEVEL set to BASIC, Oracle disables many performance statistics collections, including AWR snapshot generation.
F. They are generated if STATISTICS LEVEL is set to ALL.
True. STATISTICS_LEVEL = ALL includes all statistics collection features, so AWR snapshots are generated as well.
Summary:
AWR snapshots are normally created automatically, and they are generated when STATISTICS_LEVEL is set to TYPICAL or ALL. Therefore, the correct choices are B, D, and F.

The correct answer given is A and D, but the question asks for three steps, so there is likely a mismatch in the options or in the stated answer. Based on Oracle 19c DBCA remote PDB cloning behavior, the relevant steps are:
A. creating a database link from CDB$ROOT in the local database to CDB$ROOT in the remote system that is to be cloned
D. opening the cloned PDB
And, depending on the exact cloning method and DBCA workflow, the third step is typically not C or E as written. The most likely intended third step would be the creation of the clone itself from the source PDB, but that is not listed as an option.
Explanation of the options
A. creating a database link from CDB$ROOT in the local database to CDB$ROOT in the remote system that is to be cloned
This is correct. For a remote clone, DBCA uses a database link established from the local CDB root to the remote CDB root. This lets the local database access the remote environment and perform the clone operation.
B. creating a database link from CDB$ROOT in the local database to the PDB in the remote system that is to be cloned
This is incorrect. The database link is not typically created directly to the remote PDB for DBCA cloning. The connection is made through the remote CDB root.
C. backing up the cloned PDB
This is incorrect. Backing up the PDB is not part of the cloning steps performed by DBCA.
D. opening the cloned PDB
This is correct. After the clone is created, the new PDB is opened so it can be used. In many clone workflows, the cloned PDB is opened after creation.
E. leaving the cloned PDB in mount state
This is incorrect in this context. A newly cloned PDB is not generally left in mount state as the expected final step of the DBCA clone process.
Important note
The question appears inconsistent because it asks for three steps but only two listed options are clearly applicable. If this were an exam question, the likely intended correct choices would include A and D, plus one missing or misworded option related to creating the clone from the source PDB.
So, the best answer from the provided options is A and D, even though the “three steps” wording suggests the question or answer key may contain an error.

The correct answers are B, C, and E.
Explanation of each option:
A. All metrics are instance related.
False. Not all metrics are instance-related. Some metrics are related to the database instance, while others can be related to specific objects such as tablespaces, services, sessions, or other components.
B. Cleared stateful alerts are displayed by querying DBA_ALERT_HISTORY.
True. Stateful alerts are stored in the alert history after they have been raised and later cleared. You can query DBA_ALERT_HISTORY to see cleared stateful alerts.
C. A space usage management alert is automatically cleared after the underlying problem is resolved.
True. Space usage management alerts are stateful and will clear automatically when the condition that caused the alert, such as high tablespace usage, is no longer present.
D. They are generated by SMON when a tablespace is 97% full.
False. Alerts are not simply generated by SMON at a fixed 97% threshold. Oracle uses the server-generated alert mechanism based on thresholds and metrics, not just a hardcoded SMON event like that.
E. Metrics are statistical counts for a specific unit.
True. Metrics are measurements or statistical values collected for a specific unit of time or resource, such as a count, rate, or average.
F. STATISTICS_LEVEL must be set to ALL to generate alerts.
False. Server-generated alerts do not require STATISTICS_LEVEL to be set to ALL. Alerts are associated with metric collection and threshold monitoring, but this parameter is not a strict requirement in that way.
Summary:
The correct choices are:
B. Cleared stateful alerts are displayed by querying DBA_ALERT_HISTORY.
C. A space usage management alert is automatically cleared after the underlying problem is resolved.
E. Metrics are statistical counts for a specific unit.

The correct answers are B, C, and D.
Threshold server-generated alerts are used to notify you when a monitored database metric crosses a predefined limit. They are especially useful for detecting conditions that may indicate performance issues, space shortages, or abnormal activity.
Why B is correct:
B. when free space in a locally managed tablespace falls below a specified percentage
Threshold alerts can monitor tablespace space usage and warn you when free space gets too low. This helps you take action before you run out of space and encounter allocation failures.
Why C is correct:
C. when the total number of locked user account exceeds a specific value
Threshold alerts can be configured for certain database state metrics, including counts that reflect possible security or administrative problems. A growing number of locked accounts may indicate repeated failed logins or other access issues that need investigation.
Why D is correct:
D. when the number of logons per second exceeds a specific value
Threshold alerts can monitor activity rates such as logons per second. A sudden spike may indicate unusual workload, application issues, or even a security concern such as a login storm or unauthorized access attempts.
Why A is incorrect:
A. when free space in a dictionary managed tablespace falls below a specified percentage
Threshold alerts are associated with locally managed tablespaces, not dictionary managed tablespaces, for this kind of free space monitoring.
Why E is incorrect:
E. when a resumable statement is suspended
A resumable statement suspension is not a typical threshold-based metric alert condition. It is handled through resumable space allocation behavior and related notifications, not a threshold server-generated alert.
Why F is incorrect:
F. when the total number of logons exceeds a specific value
Threshold alerts commonly track rates, such as logons per second, rather than the total cumulative number of logons.
Summary:
The three situations where threshold server-generated alerts help diagnose and fix problems are:
B. free space in a locally managed tablespace
C. total number of locked user accounts
D. number of logons per second

The correct answers are D and E.
Why:
The problem says the read phase of a compressed RMAN incremental level 0 backup is the bottleneck while backing up to an SBT channel. That means RMAN is spending too much time reading data from the database/files before it can send it to the tape device. Since FORCE LOGGING is enabled, we should not assume any option that disables logging will help the backup read path.
Let’s examine the choices:
A. Increase the size of tape I/O buffers.
This mainly affects writing to the tape subsystem, not the read phase. Since the bottleneck is specifically in the read phase, this is not the best choice.
B. Disable FORCE LOGGING for the database.
This is unrelated to improving read performance for the backup, and it is also generally not appropriate from a data protection standpoint. FORCE LOGGING is not the cause of the read bottleneck.
C. Increase the size of the database buffer cache.
This may help normal database workload performance, but it does not directly address RMAN backup read performance in this scenario. RMAN reads data by accessing datafiles, and buffer cache size is not the key tuning lever for the backup read phase.
D. Enable asynchronous disk I/O.
Correct. Asynchronous I/O allows RMAN to issue multiple read requests without waiting for each one to complete synchronously. This can significantly improve backup read throughput, especially when reading many blocks for a compressed backup.
E. Increase the level of RMAN multiplexing.
Correct. RMAN multiplexing lets a backup piece include blocks from multiple files, which can improve read efficiency and reduce idle time in the backup process. In many RMAN backup scenarios, especially when the read phase is the bottleneck, increasing multiplexing can improve performance.
Summary:
- D improves disk read throughput through asynchronous I/O.
- E improves read efficiency by allowing RMAN to read from multiple files concurrently.
Therefore, the correct answers are D and E.

The correct answers are B and D.
The USER_TABLESPACE clause in the CREATE PLUGGABLE DATABASE command is used to control which user tablespaces are brought into a pluggable database operation. It is commonly used in scenarios where a database is being plugged in or moved, and you want to limit the tablespaces included.
Why B is correct:
B. to exclude all tablespaces except SYSTEM, SYSAUX, and TEMP when plugging in a PDB
When plugging in a PDB, the USER_TABLESPACE clause can be used to identify user tablespaces that are not to be included. This means only the required default tablespaces such as SYSTEM, SYSAUX, and TEMP remain. This matches the idea of excluding user tablespaces during a plug-in operation.
Why D is correct:
D. to specify the list of user tablespaces to include when moving a non-CDB to a PDB
When converting or moving a non-CDB into a PDB, the USER_TABLESPACE clause can be used to list the user tablespaces that should be brought over. This is a valid use of the clause because it lets you control which user tablespaces are included in the new PDB.
Why the other options are not correct:
A. to specify a default tablespace in a PDB cloned from another PDB in the same CDB
This is not the purpose of USER_TABLESPACE. Default tablespaces are configured separately and are not controlled by this clause.
C. to include specific user tablespaces only when relocating a PDB
This is not the usual syntax or use of USER_TABLESPACE. Relocation is handled differently, and this clause is not specifically for selecting tablespaces during relocation in this way.
E. to exclude a temp tablespace when plugging in a PDB
TEMP tablespaces are not handled by USER_TABLESPACE in this manner. The clause is for user tablespaces, not the temporary tablespace.
F. to specify the list of tablespaces to include when creating a PDB from the CDB seed
A PDB created from the seed uses the seed’s contents and does not use USER_TABLESPACE to select tablespaces.
Summary:
USER_TABLESPACE is used to manage user tablespaces during plug-in or conversion operations, especially when plugging in an existing database or moving a non-CDB to a PDB. Therefore, B and D are the correct choices.