cert
cert-1
cert-2

Pass Google Professional Cloud Security Engineer Exam in First Attempt Guaranteed!

Get 100% Latest Exam Questions, Accurate & Verified Answers to Pass the Actual Exam!
30 Days Free Updates, Instant Download!

cert-5
cert-6
Professional Cloud Security Engineer Exam - Verified By Experts
Professional Cloud Security Engineer Premium Bundle
$29.98

Professional Cloud Security Engineer Premium Bundle

$79.99
$109.97
  • Premium File 376 Questions & Answers. Last update: Nov 21, 2026
  • Training Course 89 Video Lectures
  • Study Guide 335 Pages
 
$109.97
$79.99
accept 28 downloads in last 7 days
block-screenshots
Purchase Google Professional Cloud Security Engineer Exam Training Products Individually
Google Professional Cloud Security Engineer Premium File
Premium File 376 Q&A
$76.99$69.99
Google Professional Cloud Security Engineer Video Training Course
Training Course 89 Lectures
$21.99 $19.99
Google Professional Cloud Security Engineer PDF Study Guide
Study Guide 335 Pages
$21.99 $19.99
Google Professional Cloud Security Engineer Practice Test 17

Question 160

A compromised application process attempts to obtain credentials from the VM metadata service. After the recommended Compute metadata exposure hardening change is implemented, which observation MOST directly verifies that it works?

  1. Encrypted identifiers follow the required format and only approved services can reverse them.
  2. Authorized workloads retrieve the needed version and revoked versions cannot be accessed.
  3. The workload retains required API access without broad metadata-derived permissions.
  4. Training jobs ingest only approved data and unauthorized dataset access fails in tests.

Explanation

Correct Answer

C:

The workload retains required API access without broad metadata-derived permissions. This directly tests whether the Compute metadata exposure hardening goal was met. Metadata server tokens reflect attached workload identity and must be protected from untrusted code paths.

Incorrect Answers

A:

Encrypted identifiers follow the required format and only approved services can reverse them. This is evidence about Format-preserving encryption data handling, not direct proof of the required Compute metadata exposure hardening behavior. The relevant check is that the workload retains required API access without broad metadata-derived permissions.

B:

Authorized workloads retrieve the needed version and revoked versions cannot be accessed. This is evidence about Secret Manager versioning and least privilege, not direct proof of the required Compute metadata exposure hardening behavior. The relevant check is that the workload retains required API access without broad metadata-derived permissions.

D:

Training jobs ingest only approved data and unauthorized dataset access fails in tests. This is evidence about AI training data isolation and least privilege, not direct proof of the required Compute metadata exposure hardening behavior. The relevant check is that the workload retains required API access without broad metadata-derived permissions.

Question 161

A support access request needs an auditable approval decision and bounded access duration. Which practical correction or implementation decision is MOST directly appropriate?

  1. Reconcile policy changes, findings, tickets and logs for the sampled time period.
  2. Review each service's responsibility boundaries against operational evidence and contracts.
  3. Verify supported service coverage, approval routing and expired request behavior.
  4. Send a helpdesk email saying approved and assume it configures the Access Approval service.

Explanation

Correct Answer

C:

Verify supported service coverage, approval routing and expired request behavior. The intended behavior can be checked because applicable requests require approved decisions and corresponding events remain traceable.

Incorrect Answers

A:

Reconcile policy changes, findings, tickets and logs for the sampled time period. This could help Audit evidence collection and attestation, but it does not resolve the Access Approval Google support access case. Access Approval governs supported Google personnel access workflows and has documented coverage limitations.

B:

Review each service's responsibility boundaries against operational evidence and contracts. This could help Cloud shared responsibility service boundaries, but it does not resolve the Access Approval Google support access case. Access Approval governs supported Google personnel access workflows and has documented coverage limitations.

D:

Send a helpdesk email saying approved and assume it configures the Access Approval service. This would not correct the Access Approval Google support access workflow. Assuming Access Approval controls every automated platform operation can misstate its scope.

Question 162

A change team requires controlled maintenance windows and success reporting for operating system patches. An engineer must explain the recommendation for VM Manager patch orchestration. Which underlying principle is the MOST accurate?

  1. Image digest attestations authorize precisely identified artifacts rather than unrelated mutable labels.
  2. Custom security findings provide evidence of drift but do not always automatically prevent resource creation.
  3. Guest OS patching remains an operational responsibility despite cloud-managed physical infrastructure.
  4. Log export requires authorized destination write paths, filters and governance beyond local logging.

Explanation

Correct Answer

C:

Guest OS patching remains an operational responsibility despite cloud-managed physical infrastructure. Therefore the implementation decision is to use supported VM Manager patch jobs and compliance views with approved maintenance scheduling.

Incorrect Answers

A:

Image digest attestations authorize precisely identified artifacts rather than unrelated mutable labels. This describes the separate Binary Authorization attestor chain of custody mechanism; the present case instead requires VM Manager patch orchestration. Untested all-fleet patching may cause application regressions or leave unpatched instances unnoticed.

B:

Custom security findings provide evidence of drift but do not always automatically prevent resource creation. This describes the separate Security Health Analytics custom finding mechanism; the present case instead requires VM Manager patch orchestration. Untested all-fleet patching may cause application regressions or leave unpatched instances unnoticed.

D:

Log export requires authorized destination write paths, filters and governance beyond local logging. This describes the separate External SIEM log export privacy mechanism; the present case instead requires VM Manager patch orchestration. Untested all-fleet patching may cause application regressions or leave unpatched instances unnoticed.

Question 163

One IP source sends thousands of login attempts to a regional application load balancer. If the relevant Cloud Armor rate limiting and bot mitigation safeguard is omitted, which outcome is the MOST directly associated failure?

  1. Allowing ingress from all subnet addresses can enable lateral movement across tiers.
  2. Treating an allow-on-443 rule as deep content validation leaves application threats unexamined.
  3. A threshold that is too low can deny legitimate users behind shared NAT addresses.
  4. Assuming a lower-priority allow overrides a higher-priority deny may leave an exposure or outage.

Explanation

Correct Answer

C:

A threshold that is too low can deny legitimate users behind shared NAT addresses. Edge rate controls mitigate abusive request volume but do not replace application authentication. The corrective approach is to validate traffic aggregation keys, request volume windows and response codes under load.

Incorrect Answers

A:

This failure pattern belongs chiefly to VPC firewall stateful filtering; it is not the failure most directly tested in the Cloud Armor rate limiting and bot mitigation situation.

B:

This failure pattern belongs chiefly to Cloud NGFW application-layer inspection; it is not the failure most directly tested in the Cloud Armor rate limiting and bot mitigation situation.

D:

This failure pattern belongs chiefly to Hierarchical Cloud NGFW policy ordering; it is not the failure most directly tested in the Cloud Armor rate limiting and bot mitigation situation.

Question 164

A development lead can deploy approved workloads but should not create unlimited privileged machine identities. If the relevant Scoped permission for service account creation safeguard is omitted, which outcome is the MOST directly associated failure?

  1. Granting sensitive cloud permissions to every node can let unrelated pods inherit excessive access.
  2. Rotating only a repository secret variable leaves an already leaked key valid.
  3. Guessing from only the project IAM screen may overlook higher-level policies.
  4. Allowing any deployer to attach an owner-level service account can enable privilege escalation.

Explanation

Correct Answer

D:

Allowing any deployer to attach an owner-level service account can enable privilege escalation. The permission to act as a service account is distinct from permission to edit the workload configuration. The corrective approach is to test deployments with permitted and forbidden target service account identities.

Incorrect Answers

A:

This failure pattern belongs chiefly to GKE workload identity binding; it is not the failure most directly tested in the Scoped permission for service account creation situation.

B:

This failure pattern belongs chiefly to Service account key inventory and exposure response; it is not the failure most directly tested in the Scoped permission for service account creation situation.

C:

Guessing from only the project IAM screen may overlook higher-level policies. This failure pattern belongs chiefly to IAM effective permission troubleshooting; it is not the failure most directly tested in the Scoped permission for service account creation situation. Allowing any deployer to attach an owner-level service account can enable privilege escalation.

Question 165

A sensitive API must deny other internal workloads even when they share the same subnet. Testing exposes this failure: Trusting all clients solely because they are inside the VPC may allow lateral impersonation. Which remediation should be applied FIRST?

  1. Use supported Cloud KMS import jobs and validate imported key version state under key policy.
  2. Test client-certificate absence, certificate rotation and authorized service policies.
  3. Separate key administration permissions from cryptographic use and decryption permissions.
  4. Reuse one unrestricted client certificate and private key across every unrelated workload.

Explanation

Correct Answer

B:

Test client-certificate absence, certificate rotation and authorized service policies. Confirmation requires seeing that both sides authenticate the expected certificate chain and unauthorized service identities are denied.

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 Mutual TLS for service-to-service identity failure. Trusting all clients solely because they are inside the VPC may allow lateral impersonation.

C:

Separate key administration permissions from cryptographic use and decryption permissions. It addresses Cloud KMS crypto key role separation, not the diagnosed Mutual TLS for service-to-service identity failure. Trusting all clients solely because they are inside the VPC may allow lateral impersonation.

D:

Reuse one unrestricted client certificate and private key across every unrelated workload. This retains the Mutual TLS for service-to-service identity problem. mTLS authenticates peers in both directions and supplements transport encryption.

Question 166

A partner needs access to approved aggregated results without base table read privileges. Select TWO statements that correctly combine the BigQuery authorized views method with evidence or its governing principle.

  1. Publish a controlled authorized view or supported approved dataset access configuration.
  2. Put a WHERE clause in a sample notebook and assume it enforces BigQuery permissions for every reader.
  3. Partner queries return only allowed columns and aggregates and direct base table queries fail.
  4. Configure Cloud SQL private connectivity and a supported database authentication model with least privilege.
  5. Store approved secrets in Secret Manager versions and grant secret access only to intended workloads.

Explanation

Correct Answers

A:

Publish a controlled authorized view or supported approved dataset access configuration. This is the intended action for BigQuery authorized views.

C:

Partner queries return only allowed columns and aggregates and direct base table queries fail. This is directly observable validation of the BigQuery authorized views result.

Incorrect Answers

B:

Put a WHERE clause in a sample notebook and assume it enforces BigQuery permissions for every reader. This is not part of a safe or effective BigQuery authorized views design. Sharing the underlying dataset viewer role can reveal more than the intended view.

D:

Configure Cloud SQL private connectivity and a supported database authentication model with least privilege. This serves the different Cloud SQL private IP and database authentication objective, not the selected BigQuery authorized views plan. Authorized views delegate access to approved query results without requiring direct base table grants.

E:

Store approved secrets in Secret Manager versions and grant secret access only to intended workloads. This serves the different Secret Manager versioning and least privilege objective, not the selected BigQuery authorized views plan. Authorized views delegate access to approved query results without requiring direct base table grants.

Question 167

An auditor needs evidence that an external key dependency was configured and monitored rather than merely promised. Which practical correction or implementation decision is MOST directly appropriate?

  1. Verify supported service coverage, approval routing and expired request behavior.
  2. Put the phrase BYOK in a bucket name and assume the service imports external keys automatically.
  3. Trace encrypted resources to key versions, regions and relevant IAM assignments.
  4. Verify sink destinations, connector scope, contracts and masked payload samples.

Explanation

Correct Answer

C:

Trace encrypted resources to key versions, regions and relevant IAM assignments. The intended behavior can be checked because key policy, authorized use and recovery tests support the stated customer custody requirement.

Incorrect Answers

A:

Verify supported service coverage, approval routing and expired request behavior. This could help Access Approval Google support access, but it does not resolve the Encryption compliance and external key governance case. Encryption compliance requires distinguishing key ownership from default provider encryption capabilities.

B:

Put the phrase BYOK in a bucket name and assume the service imports external keys automatically. This would not correct the Encryption compliance and external key governance workflow. Calling default encryption customer-owned without actual customer key management is misleading.

D:

Verify sink destinations, connector scope, contracts and masked payload samples. This could help Cross-border logs and downstream data processors, but it does not resolve the Encryption compliance and external key governance case. Encryption compliance requires distinguishing key ownership from default provider encryption capabilities.

Question 168

A managed training pipeline must read only the approved training bucket and write to a restricted model registry. Which approach BEST meets the requirement while respecting the system constraints?

  1. Enforce tool authorization and distinguish untrusted retrieved content from trusted application instructions.
  2. Attach a dedicated supported service account and restrict managed training pipeline permissions.
  3. Apply supported privacy filters, data access segmentation and output inspection for sensitive responses.
  4. Give the training VM a public JSON service account key with project Owner permissions.

Explanation

Correct Answer

B:

Attach a dedicated supported service account and restrict managed training pipeline permissions. Managed ML pipelines still depend on Google Cloud workload IAM and data protection. Verification should demonstrate that audit evidence attributes reads and writes to intended identity without unrelated dataset access.

Incorrect Answers

A:

Enforce tool authorization and distinguish untrusted retrieved content from trusted application instructions. This primarily addresses Prompt injection and retrieved data boundary, not the specific Vertex AI training and pipeline service identities decision. Here, managed ML pipelines still depend on Google Cloud workload IAM and data protection.

C:

Apply supported privacy filters, data access segmentation and output inspection for sensitive responses. This primarily addresses AI output data loss prevention, not the specific Vertex AI training and pipeline service identities decision. Here, managed ML pipelines still depend on Google Cloud workload IAM and data protection.

D:

Give the training VM a public JSON service account key with project Owner permissions. This is an unsafe or ineffective shortcut for Vertex AI training and pipeline service identities. Default or shared broad access can disclose unrelated data during experimentation.

Question 169

A team is confused because return packets are allowed despite a directional ingress rule. An engineer must explain the recommendation for VPC firewall stateful filtering. Which underlying principle is the MOST accurate?

  1. Lateral movement defenses require explicit east-west traffic restrictions independent of perimeter ingress.
  2. Private transport connectivity is distinct from cryptographic confidentiality and encrypted tunnel requirements.
  3. Access levels are contextual expressions that must be consumed by a policy to enforce a boundary.
  4. VPC firewalls evaluate directional connection initiation with stateful return traffic handling.

Explanation

Correct Answer

D:

VPC firewalls evaluate directional connection initiation with stateful return traffic handling. Therefore the implementation decision is to define least-privilege stateful VPC firewall ingress and egress controls for the intended connections.

Incorrect Answers

A:

Allowing ingress from all subnet addresses can enable lateral movement across tiers. This describes the separate N-tier east-west network segmentation mechanism; the present case instead requires VPC firewall stateful filtering.

B:

Allowing ingress from all subnet addresses can enable lateral movement across tiers. Private transport connectivity is distinct from cryptographic confidentiality and encrypted tunnel requirements.

C:

Allowing ingress from all subnet addresses can enable lateral movement across tiers. Access levels are contextual expressions that must be consumed by a policy to enforce a boundary.

Question 170

A partner must download one exported archive without receiving permanent bucket IAM membership. After the recommended Time-bounded Cloud Storage signed URLs change is implemented, which observation MOST directly verifies that it works?

  1. Authorized workloads retrieve the needed version and revoked versions cannot be accessed.
  2. Authorized applications connect privately while unrelated clients cannot reach or log into the database.
  3. Untrusted certificates fail while approved service connections negotiate authenticated encryption.
  4. The intended single-object request succeeds before expiry and fails after expiry.

Explanation

Correct Answer

D:

The intended single-object request succeeds before expiry and fails after expiry. This directly tests whether the Time-bounded Cloud Storage signed URLs goal was met. Signed URLs delegate bounded access through a signature rather than lasting IAM grants to each downloader.

Incorrect Answers

A:

Authorized workloads retrieve the needed version and revoked versions cannot be accessed. This is evidence about Secret Manager versioning and least privilege, not direct proof of the required Time-bounded Cloud Storage signed URLs behavior. The relevant check is that the intended single-object request succeeds before expiry and fails after expiry.

B:

Authorized applications connect privately while unrelated clients cannot reach or log into the database. This is evidence about Cloud SQL private IP and database authentication, not direct proof of the required Time-bounded Cloud Storage signed URLs behavior. The relevant check is that the intended single-object request succeeds before expiry and fails after expiry.

C:

Untrusted certificates fail while approved service connections negotiate authenticated encryption. This is evidence about TLS certificate trust and cipher policy, not direct proof of the required Time-bounded Cloud Storage signed URLs behavior. The relevant check is that the intended single-object request succeeds before expiry and fails after expiry.

Professional Cloud Security Engineer Practice Tests