CCSP

Free CCSP practice questions

10 questions from our CCSP practice tests, each with the correct answer and a written explanation. Read them with the answers shown, or switch to quiz mode and answer first.

Back to the CCSP practice exams

Question 1 of 10, Cloud Concepts, Architecture and Design

A government contractor encrypts archives that must stay confidential for 25 years and is concerned about future cryptographically relevant quantum computers. Which action BEST addresses this 'harvest now, decrypt later' risk?

  1. ASwitch from AES-256 to AES-128 because shorter symmetric keys are more resistant to quantum search attacks
  2. BPlan a migration to NIST-standardized post-quantum key establishment and signature algorithms with crypto agility Correct answer
  3. CRely on TLS 1.3 for archive transfers because its forward secrecy prevents any later decryption of stored data
  4. DIncrease RSA key sizes from 2048 to 4096 bits so that the archives remain secure against quantum attacks

Answer and explanation: Data needing long-term confidentiality is at risk from adversaries who capture it now and decrypt it later, so organizations should plan migration to NIST's post-quantum standards, such as ML-KEM, and build crypto agility. Larger RSA keys do not defeat Shor's algorithm. Grover's algorithm weakens symmetric keys, so AES-256 is preferred over AES-128. Forward secrecy protects past sessions from later key compromise but still relies on classical key exchange and does not protect stored archives.

Question 2 of 10, Cloud Concepts, Architecture and Design

A provider runs its compute fleet on a hypervisor installed directly on server hardware, while its developers test images locally using a hypervisor that runs as an application on their laptops' operating systems. Which statement correctly compares the two from a security perspective?

  1. AThe hosted hypervisor on laptops has the smaller attack surface because the laptop operating system shields it from guests
  2. BBoth carry an identical attack surface because each one isolates guests using the same hardware virtualization extensions
  3. CThe bare-metal hypervisor has a smaller attack surface because no general-purpose host operating system sits beneath it Correct answer
  4. DThe bare-metal hypervisor has a larger attack surface because it must include its own drivers for all server hardware

Answer and explanation: A Type 1 or bare-metal hypervisor runs directly on hardware, so there is no general-purpose host operating system whose vulnerabilities can be exploited, giving it a smaller attack surface. A Type 2 hosted hypervisor inherits every weakness of the host OS, so the host does not shield it. Shared hardware extensions do not make the overall attack surfaces equal. Bare-metal hypervisors use a minimal, purpose-built footprint rather than a general OS, so they are not considered the larger surface.

Question 3 of 10, Cloud Data Security

A payment processor stores card numbers in a cloud database. Analysts need to join transactions by card without ever seeing real numbers, and the PCI DSS assessment scope should shrink as much as possible. Which approach BEST meets both needs?

  1. AHash the card numbers with SHA-256 in the analytics copy so values can still be joined across tables
  2. BApply dynamic data masking in the database so analysts see only the last four digits of each card number
  3. CReplace card numbers with tokens from a segregated token vault so analytics systems never hold card data Correct answer
  4. DEncrypt the card column with a customer-managed key and grant analysts decrypt permission for the join queries only

Answer and explanation: Tokenization substitutes card numbers with non-sensitive tokens and keeps the mapping in a separate vault, so analytics systems that hold only tokens can be removed from PCI DSS scope while consistent tokens still support joins. Dynamic masking hides digits on display but the database still stores real card data in scope. Encryption with analyst decrypt rights keeps those systems in scope and exposes plaintext. Unsalted hashes of card numbers are vulnerable to brute force given the small keyspace, and PCI DSS still treats them cautiously.

Question 4 of 10, Cloud Data Security

A media company must ensure that leaked design files become unreadable after a partner contract ends, even if the files were already downloaded to partner laptops. Which capability is required?

  1. AClient-side encryption of the files with a key that the partner receives once at the start of the contract
  2. BData loss prevention at the egress gateway that blocks uploads of labeled files to external sites
  3. CInformation rights management with policy revocation enforced when the client next checks the license Correct answer
  4. DObject storage versioning combined with a lifecycle rule that deletes all shared copies at contract end

Answer and explanation: Information rights management keeps protection attached to the file wherever it travels and supports dynamic policy changes such as revocation, so access fails once the partner's rights are withdrawn. Egress DLP acts only at the company's network boundary and cannot reach copies already on partner devices. Deleting objects in the company's storage leaves downloaded copies untouched. A static key handed to the partner cannot be withdrawn after it has been shared.

Question 5 of 10, Cloud Platform and Infrastructure Security

An e-commerce firm runs in one cloud region across three availability zones. Leadership now requires the platform to survive the loss of an entire region with an RTO of one hour and an RPO of five minutes, at the lowest reasonable cost. Which design BEST fits?

  1. AActive-active deployment across two regions with synchronous writes and full capacity in both regions
  2. BNightly backups copied to a second region, with infrastructure-as-code to rebuild the stack on demand
  3. CSynchronous replication across the three zones with automatic failover between zones in the region
  4. DAsynchronous database replication to a second region with a scaled-down warm standby kept running Correct answer

Answer and explanation: A warm standby in a second region with asynchronous replication can achieve minutes of data loss and recovery within an hour at moderate cost. Nightly backups cannot meet a five-minute RPO, and rebuilding the whole stack is unlikely within the hour. Multi-zone replication inside one region does not survive a regional outage. Active-active with synchronous cross-region writes exceeds the requirements at much higher cost and latency.

Question 6 of 10, Cloud Platform and Infrastructure Security

A provider classifies the controls protecting its hypervisor hosts. Host intrusion detection alerts on suspicious kernel module loads, and a signed-boot chain prevents unsigned hypervisor code from starting. Which pairing of control types is correct?

  1. AIntrusion detection is corrective; the signed-boot chain is deterrent
  2. BIntrusion detection is detective, and the signed-boot chain is preventive Correct answer
  3. CIntrusion detection is compensating, and the signed-boot chain is corrective
  4. DIntrusion detection is preventive, and the signed-boot chain is detective

Answer and explanation: Intrusion detection identifies and reports activity after it starts, making it detective, while secure boot stops unauthorized code from running, making it preventive. Corrective controls restore systems after an event and deterrent controls discourage attempts. Detection does not prevent the action, and a signed-boot chain does not merely report it. A compensating control substitutes for a primary control that cannot be implemented, which is not the case here.

Question 7 of 10, Cloud Application Security

A company federates its workforce identity provider with several SaaS applications so employees sign in once with corporate credentials. Which statement BEST describes the trust relationship in this federation?

  1. AThe SaaS applications act as relying parties that accept assertions signed by the corporate identity provider Correct answer
  2. BEach SaaS application stores a copy of employee passwords and validates them against the corporate directory nightly
  3. CThe SaaS applications act as identity providers that issue authentication tokens back to the corporate directory
  4. DThe corporate directory and every SaaS application share one symmetric key that encrypts all session cookies

Answer and explanation: In federation the corporate identity provider authenticates the user and issues signed assertions or tokens, and each SaaS application is a relying party or service provider that trusts them. Copying passwords to each application defeats the purpose of federation. The roles are reversed in the option that makes SaaS apps identity providers. Federation relies on signed assertions and certificate trust, not one shared symmetric key across all applications.

Question 8 of 10, Cloud Security Operations

A company faces litigation and must identify, preserve, collect, and produce electronically stored information from several cloud services. Which standard provides guidance specific to electronic discovery?

  1. AISO/IEC 27037
  2. BISO/IEC 27041
  3. CISO/IEC 27050 Correct answer
  4. DISO/IEC 27043

Answer and explanation: ISO/IEC 27050 provides guidance on electronic discovery, covering identification, preservation, collection, processing, review, and production of electronically stored information. ISO/IEC 27037 covers identification, collection, acquisition, and preservation of digital evidence. ISO/IEC 27041 covers assurance of investigative methods. ISO/IEC 27043 covers incident investigation principles and processes.

Question 9 of 10, Cloud Security Operations

An organization deploys all cloud networking through infrastructure as code. An auditor finds a security group rule allowing database access from anywhere that does not exist in the code repository. What is the BEST long-term control?

  1. ADetect configuration drift automatically and revert or alert on resources that diverge from code Correct answer
  2. BIncrease the frequency of the manual quarterly firewall rule reviews to monthly for all production accounts
  3. CAsk administrators to document console changes in a spreadsheet maintained by the change board
  4. DDelete the rule manually and send a reminder email about the change management policy

Answer and explanation: Automated drift detection compares live resources with the declared state and reverts or alerts on unauthorized changes, enforcing infrastructure as code as the source of truth. A spreadsheet relies on voluntary reporting and does not catch unrecorded changes. Manual deletion fixes one instance but not the process gap. More frequent manual reviews still leave long exposure windows.

Question 10 of 10, Legal, Risk and Compliance

A multinational receives a litigation hold notice covering email and files stored in a SaaS suite. The suite's default retention policy deletes items after 90 days. What should the legal and security teams do FIRST?

  1. AWait for the 90-day cycle to complete so that the data set is stable before starting collection
  2. BApply a legal hold to the relevant custodians' content so it is preserved regardless of retention rules Correct answer
  3. CAsk the SaaS provider to certify in writing that no relevant data exists before taking any preservation action
  4. DExport all mailboxes to on-premises storage and then disable the SaaS retention policies entirely

Answer and explanation: Once litigation is reasonably anticipated, relevant data must be preserved immediately, and a legal hold on the custodians' content suspends deletion under normal retention. Bulk exporting everything and disabling retention is disruptive and over-broad. The provider cannot certify the customer's relevant data, and the duty to preserve stays with the customer. Letting the deletion cycle run could destroy evidence and lead to spoliation sanctions.