Google Professional Cloud Security Engineer Practice Test 29
Question 280
A regulated organization wants customer approval before applicable Google personnel access customer data for support. After the recommended Access Approval Google support access change is implemented, which observation MOST directly verifies that it works?
- Applicable requests require approved decisions and corresponding events remain traceable.
- Authorized investigators retrieve allowed aggregate values while raw patient fields remain denied.
- A sample alert is triaged, escalated and closed with attributable decision and evidence.
- Build logs show permitted operations and deny unexpected production resource access.
Explanation
Correct Answer
A:Applicable requests require approved decisions and corresponding events remain traceable. This directly tests whether the Access Approval Google support access goal was met. Access Approval governs supported Google personnel access workflows and has documented coverage limitations.
Incorrect Answers
B:Authorized investigators retrieve allowed aggregate values while raw patient fields remain denied. This is evidence about Healthcare data privacy segmentation, not direct proof of the required Access Approval Google support access behavior. The relevant check is that applicable requests require approved decisions and corresponding events remain traceable.
C:A sample alert is triaged, escalated and closed with attributable decision and evidence. This is evidence about SOC control evidence and incident workflow, not direct proof of the required Access Approval Google support access behavior. The relevant check is that applicable requests require approved decisions and corresponding events remain traceable.
D:Build logs show permitted operations and deny unexpected production resource access. This is evidence about Secure Cloud Build execution identity, not direct proof of the required Access Approval Google support access behavior. The relevant check is that applicable requests require approved decisions and corresponding events remain traceable.
Question 281
Project A peers with B and B peers with C, but A cannot reach C through B. If the relevant VPC Network Peering transitivity limits safeguard is omitted, which outcome is the MOST directly associated failure?
- Exposing a direct backend path may bypass the intended identity gate.
- Designing a three-network hub by peering alone may leave intended routes unreachable.
- A wildcard ingress policy can eliminate the intended restriction for untrusted callers.
- An IAM role alone does not confine data copied through permitted APIs to untrusted projects.
Explanation
Correct Answer
B:Designing a three-network hub by peering alone may leave intended routes unreachable. VPC Network Peering does not provide transitive routing or automatic IAM sharing. The corrective approach is to check route exchange and add the appropriate direct or supported transit architecture.
Incorrect Answers
A:Exposing a direct backend path may bypass the intended identity gate. This failure pattern belongs chiefly to Identity-Aware Proxy application access; it is not the failure most directly tested in the VPC Network Peering transitivity limits situation. Designing a three-network hub by peering alone may leave intended routes unreachable.
C:This failure pattern belongs chiefly to VPC Service Controls ingress egress policy; it is not the failure most directly tested in the VPC Network Peering transitivity limits situation.
D:This failure pattern belongs chiefly to VPC Service Controls service perimeter; it is not the failure most directly tested in the VPC Network Peering transitivity limits situation.
Question 282
An application database must not expose a public Cloud SQL endpoint to the internet. After the recommended Cloud SQL private IP and database authentication change is implemented, which observation MOST directly verifies that it works?
- Output removes intended values and controlled inspection confirms no exposed matches in sample logs.
- The intended single-object request succeeds before expiry and fails after expiry.
- Authorized applications connect privately while unrelated clients cannot reach or log into the database.
- Imported versions are usable only by authorized services and key origin records remain traceable.
Explanation
Correct Answer
C:Authorized applications connect privately while unrelated clients cannot reach or log into the database. This directly tests whether the Cloud SQL private IP and database authentication goal was met. Network connectivity and database-level authentication are separate enforcement layers.
Incorrect Answers
A:Output removes intended values and controlled inspection confirms no exposed matches in sample logs. This is evidence about Sensitive Data Protection inspection redaction, not direct proof of the required Cloud SQL private IP and database authentication behavior. The relevant check is that authorized applications connect privately while unrelated clients cannot reach or log into the database.
B:The intended single-object request succeeds before expiry and fails after expiry. This is evidence about Time-bounded Cloud Storage signed URLs, not direct proof of the required Cloud SQL private IP and database authentication behavior. The relevant check is that authorized applications connect privately while unrelated clients cannot reach or log into the database.
D:Imported versions are usable only by authorized services and key origin records remain traceable. This is evidence about Cloud KMS imported key material, not direct proof of the required Cloud SQL private IP and database authentication behavior. The relevant check is that authorized applications connect privately while unrelated clients cannot reach or log into the database.
Question 283
A regulated workload needs to reach Google APIs via internal VPC addresses rather than public egress. After the recommended Private Service Connect Google APIs endpoints change is implemented, which observation MOST directly verifies that it works?
- Access succeeds only from intended device and network contexts with policy logs explaining decisions.
- Authorized client traffic reaches the intended service privately and unrelated VPCs cannot use it by default.
- On-premises clients can reach supported APIs through the intended private Google connectivity path.
- Throughput and circuit resilience meet requirements and traffic encryption is demonstrated separately.
Explanation
Correct Answer
B:Authorized client traffic reaches the intended service privately and unrelated VPCs cannot use it by default. This directly tests whether the Private Service Connect Google APIs endpoints goal was met. Private Service Connect provides controlled private connectivity to supported services and APIs.
Incorrect Answers
A:Access succeeds only from intended device and network contexts with policy logs explaining decisions. This is evidence about Access levels with perimeter identity trust, not direct proof of the required Private Service Connect Google APIs endpoints behavior. The relevant check is that authorized client traffic reaches the intended service privately and unrelated VPCs cannot use it by default.
C:On-premises clients can reach supported APIs through the intended private Google connectivity path. This is evidence about Private Google Access for on-premises, not direct proof of the required Private Service Connect Google APIs endpoints behavior. The relevant check is that authorized client traffic reaches the intended service privately and unrelated VPCs cannot use it by default.
D:Throughput and circuit resilience meet requirements and traffic encryption is demonstrated separately. This is evidence about Cloud Interconnect transport and encryption, not direct proof of the required Private Service Connect Google APIs endpoints behavior. The relevant check is that authorized client traffic reaches the intended service privately and unrelated VPCs cannot use it by default.
Question 284
Compliance requires customer authority over key disablement for selected BigQuery tables. Testing exposes this failure: Claiming a customer controls key revocation without configuring CMEK misstates the security model. Which remediation should be applied FIRST?
- Use supported Cloud KMS import jobs and validate imported key version state under key policy.
- Inspect resource encryption configuration and authorized key-management operations.
- Use mutual TLS with approved certificate issuance and service authorization mapping.
- Disable TLS because default at-rest encryption covers all network traffic in transit.
Explanation
Correct Answer
B:Inspect resource encryption configuration and authorized key-management operations. Confirmation requires seeing that supported resources use intended key custody and policy records explain default versus customer-managed behavior.
Incorrect Answers
A:Use supported Cloud KMS import jobs and validate imported key version state under key policy. It addresses Cloud KMS imported key material, not the diagnosed Default at-rest encryption versus CMEK failure. Claiming a customer controls key revocation without configuring CMEK misstates the security model.
C:Use mutual TLS with approved certificate issuance and service authorization mapping. It addresses Mutual TLS for service-to-service identity, not the diagnosed Default at-rest encryption versus CMEK failure. Claiming a customer controls key revocation without configuring CMEK misstates the security model.
D:Disable TLS because default at-rest encryption covers all network traffic in transit. This retains the Default at-rest encryption versus CMEK problem. Default encryption at rest is provided by Google Cloud while CMEK adds customer key lifecycle authority.
Question 285
A service in a private subnet needs HTTP and HTTPS proxy filtering by destination. Which practical correction or implementation decision is MOST directly appropriate?
- Open all TCP ports to the internet and assume web destinations are restricted because a proxy exists.
- Validate issuance permissions, certificate chain and revocation or rotation processes.
- Test proxy routing, inspection policy and client fail-closed behavior under outage.
- Inventory forwarding rules, NAT paths, routes and service endpoint exposure.
Explanation
Correct Answer
C:Test proxy routing, inspection policy and client fail-closed behavior under outage. The intended behavior can be checked because allowed destinations succeed, blocked hosts fail and outbound proxy logs identify decisions.
Incorrect Answers
A:Open all TCP ports to the internet and assume web destinations are restricted because a proxy exists. This would not correct the Secure Web Proxy outbound web egress workflow. Bypassing the proxy with direct egress defeats approved destination filtering.
B:Validate issuance permissions, certificate chain and revocation or rotation processes. This could help Certificate Authority Service private PKI, but it does not resolve the Secure Web Proxy outbound web egress case. A web proxy enforces application-aware outbound web policy rather than serving as inbound load balancing.
D:Inventory forwarding rules, NAT paths, routes and service endpoint exposure. This could help Private versus public endpoint design, but it does not resolve the Secure Web Proxy outbound web egress case. A web proxy enforces application-aware outbound web policy rather than serving as inbound load balancing.
Question 286
Security wants to identify legitimate dependency traffic before enforcing restrictions. Which practical correction or implementation decision is MOST directly appropriate?
- Review BGP route advertisements and test the permitted hybrid flow matrix.
- Treat a dry-run violation log as proof the API request was actually blocked.
- Diagnose deny logs and IAM policy evaluation independently for the exact principal.
- Analyze dry-run denial logs and iteratively approve only required flows before enforcement.
Explanation
Correct Answer
D:Analyze dry-run denial logs and iteratively approve only required flows before enforcement. The intended behavior can be checked because logs identify would-be denied API calls while existing workloads continue to operate.
Incorrect Answers
A:Review BGP route advertisements and test the permitted hybrid flow matrix. This could help Hybrid network segmentation policy placement, but it does not resolve the VPC Service Controls dry-run evaluation case. VPC Service Controls dry-run reports policy violations without blocking them.
B:Treat a dry-run violation log as proof the API request was actually blocked. This would not correct the VPC Service Controls dry-run evaluation workflow. Enforcing an untested perimeter can cause production data pipeline failures.
C:Diagnose deny logs and IAM policy evaluation independently for the exact principal. This could help VPC Service Controls versus IAM authorization, but it does not resolve the VPC Service Controls dry-run evaluation case. VPC Service Controls dry-run reports policy violations without blocking them.
Question 287
A change team requires controlled maintenance windows and success reporting for operating system patches. Which practical correction or implementation decision is MOST directly appropriate?
- Change the VM label to patched and assume security updates have been installed.
- Test revision deployment, job execution and documented emergency break-glass logging.
- Test write, delete and pull behavior for several repository-specific identities.
- Pilot patch deployment, verify reboot policy and collect patch job evidence.
Explanation
Correct Answer
D:Pilot patch deployment, verify reboot policy and collect patch job evidence. The intended behavior can be checked because the selected instances install approved updates and show patch compliance and failures.
Incorrect Answers
A:Change the VM label to patched and assume security updates have been installed. This would not correct the VM Manager patch orchestration workflow. Untested all-fleet patching may cause application regressions or leave unpatched instances unnoticed.
B:Test revision deployment, job execution and documented emergency break-glass logging. This could help Binary Authorization Cloud Run revision policy, but it does not resolve the VM Manager patch orchestration case. Guest OS patching remains an operational responsibility despite cloud-managed physical infrastructure.
C:Test write, delete and pull behavior for several repository-specific identities. This could help Artifact Registry repository IAM isolation, but it does not resolve the VM Manager patch orchestration case. Guest OS patching remains an operational responsibility despite cloud-managed physical infrastructure.
Question 288
An automation service maintains a security group for a rotating set of approved analysts. After the recommended Programmatic group membership governance change is implemented, which observation MOST directly verifies that it works?
- Membership changes match approved source records and unexpected external identities are rejected.
- Sign-in and API scopes are verified separately according to their respective trust boundaries.
- The scheduled synchronization creates and updates the intended users without unexpectedly deleting cloud identities.
- Emergency accounts can regain administrative access during an identity-provider outage without being used routinely.
Explanation
Correct Answer
A:Membership changes match approved source records and unexpected external identities are rejected. This directly tests whether the Programmatic group membership governance goal was met. Managing authorization through groups can simplify changes but group administration itself is privileged.
Incorrect Answers
B:Sign-in and API scopes are verified separately according to their respective trust boundaries. This is evidence about SAML versus OAuth authentication purpose, not direct proof of the required Programmatic group membership governance behavior. The relevant check is that membership changes match approved source records and unexpected external identities are rejected.
C:The scheduled synchronization creates and updates the intended users without unexpectedly deleting cloud identities. This is evidence about Cloud Identity directory synchronization, not direct proof of the required Programmatic group membership governance behavior. The relevant check is that membership changes match approved source records and unexpected external identities are rejected.
D:Emergency accounts can regain administrative access during an identity-provider outage without being used routinely. This is evidence about Super administrator emergency access, not direct proof of the required Programmatic group membership governance behavior. The relevant check is that membership changes match approved source records and unexpected external identities are rejected.
Question 289
A team must audit which service identities invoked its managed AI endpoint during a sensitive incident. Testing exposes this failure: An open inference endpoint can expose private prompts or generate unauthorized costs. Which remediation should be applied FIRST?
- Publish an unauthenticated model endpoint and rely on an obscure URL as the only access control.
- Classify responsibility for infrastructure-managed versus customer-managed AI deployment components.
- Test permitted and denied endpoint invocation with redacted telemetry.
- Apply supported privacy filters, data access segmentation and output inspection for sensitive responses.
Explanation
Correct Answer
C:Test permitted and denied endpoint invocation with redacted telemetry. Confirmation requires seeing that only approved workload principals can invoke the model and usage audit events are attributable.
Incorrect Answers
A:Publish an unauthenticated model endpoint and rely on an obscure URL as the only access control. This retains the Managed AI endpoint access controls problem. Model inference endpoints are APIs that still require authorization, quotas and protected logging.
B:Classify responsibility for infrastructure-managed versus customer-managed AI deployment components. It addresses IaaS versus managed AI training boundary, not the diagnosed Managed AI endpoint access controls failure. An open inference endpoint can expose private prompts or generate unauthorized costs.
D:Apply supported privacy filters, data access segmentation and output inspection for sensitive responses. It addresses AI output data loss prevention, not the diagnosed Managed AI endpoint access controls failure. An open inference endpoint can expose private prompts or generate unauthorized costs.
Question 290
A healthcare support log contains accidental patient email addresses in free text. Which approach BEST meets the requirement while respecting the system constraints?
- Configure supported deterministic tokenization or cryptographic de-identification with restricted key access.
- Migrate to uniform bucket-level access and enforce bucket IAM on eligible object operations.
- Apply Sensitive Data Protection de-identification transforms that redact selected infoTypes.
- Use a public regex that matches only one test phone number and certify all patient data sanitized.
Explanation
Correct Answer
C:Apply Sensitive Data Protection de-identification transforms that redact selected infoTypes. Redaction obscures selected content but requires correct detector configuration and validation. Verification should demonstrate that output removes intended values and controlled inspection confirms no exposed matches in sample logs.
Incorrect Answers
A:Configure supported deterministic tokenization or cryptographic de-identification with restricted key access. This primarily addresses Sensitive Data Protection deterministic de-identification, not the specific Sensitive Data Protection inspection redaction decision. Here, redaction obscures selected content but requires correct detector configuration and validation.
B:Migrate to uniform bucket-level access and enforce bucket IAM on eligible object operations. This primarily addresses Cloud Storage uniform bucket-level permissions, not the specific Sensitive Data Protection inspection redaction decision. Here, redaction obscures selected content but requires correct detector configuration and validation.
D:Use a public regex that matches only one test phone number and certify all patient data sanitized. This is an unsafe or ineffective shortcut for Sensitive Data Protection inspection redaction. Overly broad or incomplete detectors can destroy useful logs or leave sensitive fields exposed.