CISSP (Certified Information Systems Security Professional) — All Questions

In a system that enforces the Bell-LaPadula model, a process running at Secret attempts to write data into an Unclassified file. Which property blocks the write?

• a.The tranquility property
• b.The Biba integrity property
• c.The *-property✓
• d.The simple security property

RFC 4949 defines the confinement (*-) property as allowing write access only if the object's classification dominates the subject's clearance, so writing down from Secret to Unclassified is blocked to prevent leakage. The simple security property governs reading, tranquility concerns security levels not changing during processing, and Biba is a separate integrity model.

A user cleared for Confidential attempts to read a Secret document on a system enforcing Bell-LaPadula. Which rule denies the request?

• a.A Clark-Wilson certification rule
• b.The Brewer-Nash read rule
• c.The simple security property✓
• d.The *-property (confinement property)

RFC 4949 defines the simple security property as allowing read access only if the subject's clearance dominates the object's classification, so a Confidential user cannot read up to Secret. The *-property governs writing, Clark-Wilson is a commercial integrity model, and the Brewer-Nash rule addresses conflicts of interest between firms rather than clearance levels.

A laboratory system must prevent data from low-integrity sensors from being written into its high-integrity calibration records. Which model is designed for this goal?

• a.Biba✓
• b.Brewer-Nash
• c.Bell-LaPadula
• d.Take-Grant

RFC 4949 describes the Biba model as an integrity model in which each subject and object has an integrity level and a subject may not change information in an object at a higher or incomparable level. Bell-LaPadula protects confidentiality, Brewer-Nash enforces a Chinese wall against conflicts of interest, and Take-Grant models how access rights propagate.

A consultant who has read confidential files for Bank A is later blocked from opening files for Bank B, a competitor, but may still open files for an unrelated retailer. Which security model does this behavior implement?

• a.Biba
• b.Brewer-Nash✓
• c.Bell-LaPadula
• d.Clark-Wilson

RFC 4949 describes the Brewer-Nash model as enforcing the Chinese wall policy: a subject may read an object only if it is from a firm already accessed or belongs to a conflict-of-interest class the subject has not yet read from. Bell-LaPadula and Biba use fixed clearance and integrity levels, and Clark-Wilson focuses on commercial data integrity through controlled transactions.

A firewall appliance suffers a software fault. The design requires it to stop forwarding traffic rather than pass everything through unfiltered. Which design principle is being applied?

• a.Privacy by design
• b.Keep it simple and small
• c.Shared responsibility
• d.Fail securely✓

NIST SP 800-53 Rev. 5 control SC-24 requires components to fail to an organization-defined known state so that failures do not cause loss of confidentiality, integrity or availability, which the ISC2 outline lists as the fail-securely principle. Simplicity reduces attack surface, shared responsibility divides duties between a cloud provider and customer, and privacy by design embeds privacy protections; none of them defines behavior on failure.

A company moves to a zero trust architecture. Which assumption does NIST SP 800-207 reject?

• a.Access decisions can use device health and behavior
• b.Resources should be protected at a granular level
• c.Devices on the internal network can be trusted by location✓
• d.Every access request should be authenticated and authorized

NIST SP 800-207 states that zero trust assumes there is no implicit trust granted to assets or user accounts based solely on their physical or network location. Authenticating and authorizing each request, using device state and behavior in dynamic policy, and protecting resources individually are all tenets that SP 800-207 endorses.

In the NIST SP 800-207 zero trust logical architecture, which component makes and logs the decision to grant, deny or revoke access to a resource?

• a.The policy engine✓
• b.The public key infrastructure
• c.The policy enforcement point
• d.The data access policy store

SP 800-207 splits the policy decision point into the policy engine, which makes and logs the access decision, and the policy administrator, which executes it by setting up or shutting down the communication path. The policy enforcement point enables, monitors and terminates connections as instructed, while policy stores and PKI are supporting data sources, not decision makers.

A company runs its web servers on infrastructure as a service (IaaS). Which task remains the customer's responsibility under the NIST definition of IaaS?

• a.Controlling physical access to the provider's facility
• b.Replacing failed physical disks in the data center
• c.Maintaining the hypervisor on the provider's hosts
• d.Patching the guest operating systems✓

NIST SP 800-145 states that the IaaS consumer does not manage the underlying cloud infrastructure but has control over operating systems, storage and deployed applications, so guest OS patching stays with the customer. Physical disks, the virtualization layer and facility security are part of the underlying infrastructure the provider manages.

An organization adopts a software as a service (SaaS) email platform. Under NIST SP 800-145, what can the customer typically still configure?

• a.The provider's server operating systems
• b.The storage arrays behind the service
• c.Limited user-specific application settings✓
• d.The network layout of the provider's data centers

NIST SP 800-145 states that the SaaS consumer does not manage or control the underlying network, servers, operating systems, storage or even individual application capabilities, with the possible exception of limited user-specific application configuration settings. Operating systems, storage and network design all belong to the provider under this model.

An architect specifies AES for a new storage system. According to FIPS 197, which block and key sizes does AES use?

• a.64-bit blocks; 56-, 112- or 168-bit keys
• b.A block size that varies with the key length
• c.256-bit blocks; 128-bit keys only
• d.128-bit blocks; 128-, 192- or 256-bit keys✓

FIPS 197 states that AES can use keys of 128, 192 and 256 bits to encrypt and decrypt data in blocks of 128 bits, and AES-128, AES-192 and AES-256 all use that same block size. 64-bit blocks belong to DES and Triple DES, and although Rijndael supports other block sizes, the AES standard fixes the block at 128 bits.

A software publisher wants customers to confirm that a downloaded installer came from the publisher and was not modified. What should the publisher do?

• a.Publish the installer's hash on the same download page only
• b.Sign the installer with its private key✓
• c.Compress the installer with a password
• d.Encrypt the installer with the customer's public key

A digital signature made with the publisher's private key and checked with its public key proves origin and integrity; NIST SP 800-218 task PS.2.1 calls for making software integrity verification information available to acquirers. Encrypting with a customer's public key provides confidentiality to one recipient, not proof of origin; a hash posted on the same compromised page can be replaced along with the file; and a password-protected archive does not prove who created it.

A team must send a large file to a partner securely and efficiently using public key infrastructure. Which approach is standard practice?

• a.Encrypt the whole file with the partner's public key
• b.Encrypt the file with the team's own private key
• c.Hybrid encryption: symmetric data key, wrapped with the partner's public key✓
• d.Encrypt the file with a key both sides derived from a shared password

Asymmetric algorithms are slow and suited to transporting keys, so NIST SP 800-57 Part 1 describes key transport in which a symmetric data-encryption key is protected with the recipient's public key while the bulk data is encrypted symmetrically. Encrypting everything with a public key is inefficient, 'encrypting' with one's own private key is a signature operation anyone can reverse with the public key, and a password-derived key is not a PKI solution.

Which concept does NIST SP 800-57 use to describe the time span during which a specific key is authorized for use?

• a.Key escrow
• b.Key clustering
• c.Key stretching
• d.Cryptoperiod✓

SP 800-57 Part 1 defines a cryptoperiod as the time span during which a specific key is authorized for use by legitimate entities, and limiting it limits the data exposed if a key is compromised. Key escrow, key stretching and key clustering are other key-related concepts; none of them is the authorized usage period of a key.

A server certificate's private key is stolen three months before the certificate expires. How do relying parties learn that the certificate is no longer valid?

• a.The CA revokes it and publishes its status✓
• b.The server stops sending the certificate chain
• c.The certificate expires early on its own
• d.Browsers detect the theft from the key size

RFC 5280 describes circumstances such as key compromise that cause a certificate to become invalid before its validity period ends, and the CA then revokes it and publishes that status through a certificate revocation list (or an online status service). Certificates do not change their own validity dates, withholding the chain does not stop an attacker using the stolen key, and key size reveals nothing about theft.

Attackers steal a password database and try to use precomputed tables of hashes against it. Which control most directly defeats precomputed tables?

• a.A unique random salt per password✓
• b.Requiring passwords to expire every 30 days
• c.Encrypting the database backup tapes
• d.Hashing passwords with a faster algorithm

RFC 4949 explains that concatenating each password with its own random salt before applying the one-way function increases the difficulty of off-line dictionary attacks, because a value precomputed for one entry cannot be reused against another. Encrypted backups do not protect a stolen live database, periodic expiry does not stop offline cracking of what was stolen (and SP 800-63B-4 forbids forcing it), and a faster hash only makes precomputation and guessing cheaper.

Why do organizations plan migration to the post-quantum algorithms standardized in FIPS 203, 204 and 205?

• a.Classical computers can now factor 2048-bit RSA keys
• b.Quantum computers make AES-256 decryptable in seconds
• c.Hash functions such as SHA-256 have been proven broken
• d.Quantum computers could break RSA and elliptic curve✓

NIST released FIPS 203 (ML-KEM), FIPS 204 (ML-DSA) and FIPS 205 (SLH-DSA) because a cryptographically relevant quantum computer could break the widely deployed public-key algorithms based on factoring and discrete logarithms, including RSA and elliptic curve cryptography. FIPS 203 places the quantum risk on public-key schemes based on factoring and discrete logarithms; it makes no claim that AES-256 falls in seconds, that classical computers now factor 2048-bit RSA, or that SHA-256 is broken.