Select Two (Multi-Select) Questions

Number of database objects

Each database object must be processed during upgrade scripts, so a larger object count lengthens the upgrade window. The number of objects is a direct driver of upgrade duration, alongside data volume and dependent object complexity, making it a genuine influencing factor.

System resources (CPU/IO)

CPU and IO throughput determine how quickly upgrade scripts, dictionary conversions and index rebuilds complete. The stem asks which factors influence upgrade duration; constrained system resources directly lengthen each upgrade phase, making hardware capacity a determining constraint.

Database size

Larger databases require more datafile copying, block migration and index rebuilding during upgrade, directly extending elapsed time. The stem asks which factors influence upgrade duration; database size is a primary determinant because the volume of data to convert scales the work performed.

Exporting the recovered table to the target database using Data Pump

RECOVER TABLE uses Data Pump to export the recovered table from the auxiliary instance and import it into the target database, restoring the table's data and metadata. This satisfies the requirement that the table be returned to its original schema automatically.

Restoration of the tablespace containing the table to an auxiliary location

RMAN restores the tablespace containing the dropped table to an auxiliary location, not the production datafiles, so the target database remains untouched during recovery. This satisfies the constraint of recovering a single table without affecting the live tablespace.

Creation of an auxiliary instance

RMAN automatically creates an auxiliary instance to perform table recovery, satisfying the requirement that recovery occurs without affecting the production database. This auxiliary instance mounts the necessary backups, applies redo to a point before the table's modification, then exports the table into a dump file for import into the target.

Datafile locations

The PDB XML metadata file records datafile locations so the PDB can be cloned or plugged in elsewhere with correct file paths. This satisfies the requirement by capturing the physical storage layout needed for relocation.

PDB GUID

The PDB XML metadata file stores the PDB GUID, the globally unique identifier assigned at creation, which remains constant across plug operations. This satisfies the requirement by providing the identity information needed to track the PDB.

Character set

The PDB XML metadata file records the database character set and national character set, ensuring the plugged-in PDB remains compatible with its target CDB. This satisfies the requirement by preserving character set information across plug operations.

Library cache statistics

Important for parsing/memory health.

System-wide wait events

Provides overview of system health.

SQL execution statistics

Crucial for identifying heavy queries.

Manual addition of optimizer hints

Optimizer hints embedded in the SQL text directly override the optimizer's cost-based decisions, forcing a specific join method, index or access path. This manual method gives the DBA explicit control when statistics or transformations produce a poor plan.

SQL Tuning Advisor

SQL Tuning Advisor analyses a statement, its statistics and execution history, then produces recommendations such as SQL Profiles, indexes, restructured SQL or statistics gathering. It automates diagnosis, satisfying the stem's requirement for a tuning method beyond manual hint insertion.

Update the connection pool

The application's connection pool must be updated so it can route read-only connections to the True Cache instance rather than only the primary database. This satisfies the application-tier routing requirement, directing eligible queries to the cache.

Configure the True Cache instance

A True Cache instance must be created and configured on its own host, with its parameters and primary database association set. This satisfies the prerequisite of having a running cache instance before applications can route read-only queries to it.

Enable on the application side

True Cache must be enabled on the application side, typically by setting the connection property that marks connections as read-only cache-eligible. This satisfies the requirement that the application explicitly opt in to using True Cache.

AWR Report

AWR reports capture persistent, historical snapshots of database workload statistics, enabling comparison across days or weeks. This directly satisfies the long-term trend requirement, since the stem asks for diagnosing performance over time rather than instantaneous activity. Unlike real-time views, AWR retains data in the workload repository for retrospective analysis.

ADDM Report

ADDM reports analyse past performance by comparing consecutive Automatic Workload Repository snapshots, identifying the highest-impact issues and their root causes. This directly satisfies the stem's long-term trend requirement, since ADDM retains historical findings across multiple analysis periods rather than showing only current activity.

Reviewing log files

Upgrade scripts and the Data Pump or catupgrd logs record each component's execution, including errors and invalid objects. Reviewing them satisfies the stem's verification requirement by exposing failures that a version banner alone would conceal, confirming every component reached the expected state.

Querying V$VERSION

Querying V$VERSION returns the active banner for the running instance, so it directly confirms the database has restarted into the new release. This satisfies the stem's requirement to verify upgrade success by proving the target version is genuinely in effect, not merely installed on disk.

V$SESSION_WAIT

V$SESSION_WAIT exposes the current wait event, wait time and wait parameters for each active session, letting you pinpoint which sessions are stalled and on what resource right now. This directly satisfies the requirement to monitor current session performance rather than historical trends.

V$SESSION

V$SESSION exposes one row per active session, including live wait events, blocking session identifiers and SQL execution statistics, so it reports current session performance rather than historical data. This directly satisfies the requirement to monitor sessions as they are running, unlike views sourced from AWR or Statspack.

VERIFY

VERIFY is permitted on Blockchain tables, returning a cryptographic proof that rows remain untampered since insertion. It satisfies the stem's requirement by confirming ledger integrity without altering data, unlike DML operations that Blockchain tables restrict. Oracle implements this through hash-chain validation across the table's historical row versions.

INSERT

INSERT appends new rows, each carrying a hash chained to the previous row and signed with the table's signing key. This is permitted because Blockchain tables are append-only, so rows can be added but never updated or deleted.

SELECT

SELECT reads rows from a blockchain table without altering the tamper-evident chain, satisfying the stem's requirement for permitted operations. Blockchain tables in Oracle Database restrict DML to INSERT only; UPDATE and DELETE are blocked, while SELECT and indexing remain allowed. Reading therefore preserves the cryptographic row hashes that secure the ledger.

High resource contention

High resource contention directly degrades performance because CPU, memory, disk, and network capacity are finite; when multiple processes compete for the same constrained resource, queuing delays increase and throughput falls. This satisfies the stem's requirement for a common degradation source, as contention predictably slows response times across workloads.

Inefficient SQL statements

Inefficient SQL statements force full table scans and excessive logical reads, saturating disk I/O and CPU on the database server. This directly satisfies the stem's demand for a common performance-degradation source, since poorly written queries remain one of the most frequent causes of slow application response in production systems.

Recompile invalid objects

Recompiling invalid objects rebuilds PL/SQL units and views whose dependencies changed during the upgrade, since schema modifications invalidate stored code. This restores executable status and correct dependency chains, a required post-upgrade task before the database can serve application workloads reliably.

Run timezone upgrade script

After an Oracle upgrade, the timezone data files must be refreshed to match the new database version. Running the timezone upgrade script updates the bundled timezone definitions, preventing timestamp and scheduling errors in applications relying on region-specific time data.

Gather dictionary statistics

Gathering dictionary statistics after upgrade refreshes the optimiser's data dictionary metadata, which becomes stale because the upgrade modifies schema objects. This restores accurate cardinality estimates for the new release's execution plans, a mandatory post-upgrade step before validating workload performance.

Existing lock files

Indicates another process is running.

Insufficient space

Critical prerequisite.

Incorrect permissions

Prevents execution.

Creating a SQL profile

A SQL profile supplies auxiliary statistics and hints that steer the optimizer toward a better plan without changing the SQL text or schema objects. SQL Tuning Advisor recommends this when plan regressions or misestimates persist, directly improving performance.

Creating a new index

SQL Tuning Advisor examines access paths and can recommend creating a new index when it detects full table scans or inefficient joins, directly addressing the stem's performance-improvement requirement. This is a distinct recommendation class from statistics refresh and SQL profile creation.

Updating optimizer statistics

Refreshing stale or missing dictionary statistics directly addresses the root cause when the optimiser misestimates cardinality and selects a poor plan. The SQL Tuning Advisor routinely recommends gathering statistics on tables, indexes or columns referenced by the analysed statement, satisfying the stem's requirement for a genuine performance recommendation.

Background processes

Background processes such as DBWn, LGWR, CKPT and SMON belong to the CDB instance and serve every PDB; PDBs have no separate processes. This shared instance-level architecture is what distinguishes a CDB from non-CDB deployments.

Redo log files

Redo log files belong to the CDB instance rather than any individual PDB, so all pluggable databases share the same redo stream for instance recovery. This differs from undo tablespaces and data dictionary objects, which are per-PDB.

SGA

The SGA is allocated once by the CDB instance and shared by all PDBs, since PDBs are logical containers without their own memory structures. Each PDB draws from the same buffer cache and shared pool.

Post-upgrade cleanup

AutoUpgrade includes post-upgrade cleanup, removing obsolete artefacts and invalid objects left behind after the database conversion. This satisfies the stem's requirement for a genuine AutoUpgrade action, as the tool performs remediation steps after the upgrade completes rather than only pre-checks or the upgrade itself.

Upgrading the database

AutoUpgrade performs the actual database upgrade as its core action, converting the pre-upgrade environment to the target release. This satisfies the stem's requirement for a genuine AutoUpgrade phase, since the tool's central purpose is executing the upgrade rather than merely analysing or cleaning up afterwards.

Pre-check analysis

AutoUpgrade runs pre-check analysis before conversion, scanning the source database for conditions that would block or complicate the upgrade. This satisfies the stem's requirement for a genuine AutoUpgrade action, since the tool's first phase is diagnostic assessment rather than the upgrade or cleanup itself.

They are stored in the Fast Recovery Area (FRA).

Flashback logs reside in the Fast Recovery Area, the configured disk location Oracle uses for recovery-related files. This placement satisfies the storage constraint, since the FRA is where the database writes flashback data used by Flashback Database operations.

They are circular in nature.

Flashback logs are reused circularly, with the oldest entries overwritten once space is exhausted, satisfying the characteristic of circular behaviour. This differs from archived redo logs, which are retained until explicitly deleted or made obsolete by a retention policy.

They are used to revert the database state.

Flashback logs let you rewind the database to a prior point in time, satisfying the requirement to revert database state. FLASHBACK DATABASE applies them to return datafiles to an earlier SCN without restoring from backup, reversing unwanted changes.

Dedicated local users

Dedicated local users exist only within a single pluggable database, not in the root container, satisfying the requirement for characteristics unique to PDBs. Unlike common users, who are created in the root and shared across all containers, local users are scoped strictly to their own PDB and cannot authenticate elsewhere in the multitenant architecture.

Independent datafiles

Independent datafiles let each pluggable database own its own tablespaces and operating system files, so one PDB's storage can be relocated, backed up or unplugged without touching others. This satisfies the stem's requirement for a characteristic unique to PDBs, since a non-CDB has no such per-database file isolation within a shared container.

Individual services

Individual services let each PDB register its own service name with the listener, so clients connect directly to a specific pluggable database. This per-PDB service registration is unique to the multitenant architecture, unlike a non-CDB instance's single service.

Optimizer statistics

Optimiser statistics describe table, column and index data distribution, feeding cardinality estimates that drive cost calculations. They satisfy the stem because inaccurate or stale statistics cause the optimiser to select a plan based on wrong row-count assumptions.

Indexes on the queried columns

Available indexes determine which access paths the optimiser can cost, influencing whether it chooses an index scan or full table scan. They satisfy the stem because the presence or absence of a suitable index directly changes the cheapest plan considered.

Initialization parameters

Initialization parameters set optimiser behaviour and resource limits at instance level, including optimiser_mode and cursor_sharing, which directly shape plan selection. They satisfy the stem because these settings constrain the cost model the optimiser applies before choosing an execution plan.

It resides in the application tier

True Cache runs as a separate Oracle instance in the application tier, close to the application servers rather than alongside the primary database. This placement reduces network round-trip latency for read queries while the primary database remains the single source of truth.

It is a read-only cache

True Cache is strictly read-only: it serves queries but never accepts writes. All DML is routed to the primary database, and the cache refreshes itself automatically from the primary, keeping data current without any application-side write logic.