Google Professional Cloud Security Engineer Practice Test 1
Question 1
A support access request needs an auditable approval decision and bounded access duration. Select TWO statements that correctly combine the Access Approval Google support access method with evidence or its governing principle.
- Send a helpdesk email saying approved and assume it configures the Access Approval service.
- Configure supported Access Approval settings and authorized approvers for eligible services.
- Applicable requests require approved decisions and corresponding events remain traceable.
- Review available Access Transparency logs alongside Cloud Audit Logs for the appropriate services.
- Map controls to documented incident runbooks, test outcomes and preserved audit logs.
Explanation
Correct Answers
B:Configure supported Access Approval settings and authorized approvers for eligible services. This is the intended action for Access Approval Google support access.
C:Applicable requests require approved decisions and corresponding events remain traceable. This is directly observable validation of the Access Approval Google support access result.
Incorrect Answers
A:Send a helpdesk email saying approved and assume it configures the Access Approval service. This is not part of a safe or effective Access Approval Google support access design. Assuming Access Approval controls every automated platform operation can misstate its scope.
D:Review available Access Transparency logs alongside Cloud Audit Logs for the appropriate services. This serves the different Access Transparency provider audit visibility objective, not the selected Access Approval Google support access plan. Access Approval governs supported Google personnel access workflows and has documented coverage limitations.
E:Map controls to documented incident runbooks, test outcomes and preserved audit logs. This serves the different SOC control evidence and incident workflow objective, not the selected Access Approval Google support access plan. Access Approval governs supported Google personnel access workflows and has documented coverage limitations.
Question 2
A security engineer needs to deactivate one suspicious service account during investigation. An engineer must explain the recommendation for Service account disabling and dependency mapping. Which underlying principle is the MOST accurate?
- GKE workload identity separates pod-level resource authorization from the node's broad execution identity.
- Phishing-resistant authenticators bind verification to the legitimate service origin.
- Disabling an identity affects all callers using it, not just one label or attached role.
- Long-lived service account keys are exportable credentials that require strict inventory and incident response.
Explanation
Correct Answer
C:Disabling an identity affects all callers using it, not just one label or attached role. Therefore the implementation decision is to inventory workloads using the identity and disable or suspend the affected service account with a recovery plan.
Incorrect Answers
A:Blindly deleting a shared service account can cause unplanned production outages. This describes the separate GKE workload identity binding mechanism; the present case instead requires Service account disabling and dependency mapping.
B:Blindly deleting a shared service account can cause unplanned production outages. This describes the separate Phishing-resistant multi-factor verification mechanism; the present case instead requires Service account disabling and dependency mapping.
D:Blindly deleting a shared service account can cause unplanned production outages. This describes the separate Service account key inventory and exposure response mechanism; the present case instead requires Service account disabling and dependency mapping.
Question 3
An auditor needs a clear boundary separating in-scope workloads from unrelated application environments. An engineer must explain the recommendation for Regulatory scope inventory and boundaries. Which underlying principle is the MOST accurate?
- AI training workloads inherit ordinary cloud identity and data governance responsibilities.
- Scoping is based on actual data and connected system boundaries rather than arbitrary project names.
- Assured Workloads provides configurable compliance guardrails but does not certify arbitrary workloads by mere association.
- Data classification drives control selection and should be based on sensitivity and legal purpose.
Explanation
Correct Answer
B:Scoping is based on actual data and connected system boundaries rather than arbitrary project names. Therefore the implementation decision is to document in-scope resources, identities, services, networks, external dependencies and data flows.
Incorrect Answers
A:Missing a shared logging or integration dependency can understate the regulated environment. AI training workloads inherit ordinary cloud identity and data governance responsibilities.
C:Missing a shared logging or integration dependency can understate the regulated environment. Assured Workloads provides configurable compliance guardrails but does not certify arbitrary workloads by mere association.
D:Missing a shared logging or integration dependency can understate the regulated environment. This describes the separate Sensitive data classification and policy labels mechanism; the present case instead requires Regulatory scope inventory and boundaries.
Question 4
A private administrator website needs staff SSO rather than public password-based internet exposure. After the recommended Identity-Aware Proxy application access change is implemented, which observation MOST directly verifies that it works?
- Cross-perimeter attempts to restricted services are denied while required in-perimeter workflows work.
- Signed-in authorized users reach the app and rejected identities cannot bypass the IAP path.
- Restricted clients receive the expected filtered response without affecting unrelated resolvers.
- Only intended networks or authenticated clients reach the workload while public scans fail.
Explanation
Correct Answer
B:Signed-in authorized users reach the app and rejected identities cannot bypass the IAP path. This directly tests whether the Identity-Aware Proxy application access goal was met. IAP adds identity-based access enforcement before reaching supported applications.
Incorrect Answers
A:Cross-perimeter attempts to restricted services are denied while required in-perimeter workflows work. This is evidence about VPC Service Controls service perimeter, not direct proof of the required Identity-Aware Proxy application access behavior. The relevant check is that signed-in authorized users reach the app and rejected identities cannot bypass the IAP path.
C:Restricted clients receive the expected filtered response without affecting unrelated resolvers. This is evidence about Cloud DNS response policy filtering, not direct proof of the required Identity-Aware Proxy application access behavior. The relevant check is that signed-in authorized users reach the app and rejected identities cannot bypass the IAP path.
D:Only intended networks or authenticated clients reach the workload while public scans fail. This is evidence about Private versus public endpoint design, not direct proof of the required Identity-Aware Proxy application access behavior. The relevant check is that signed-in authorized users reach the app and rejected identities cannot bypass the IAP path.
Question 5
Regulators require supported storage resources to comply with a company-specific configuration predicate. After the recommended Custom organization policy on resource attributes change is implemented, which observation MOST directly verifies that it works?
- Unapproved external principals cannot be added to controlled IAM policies.
- Compliant test creations succeed and prohibited configurations are rejected with useful error evidence.
- Test users complete sign-in through the designated provider and break-glass access remains available.
- Unapproved VMs cannot acquire public IP addresses while approved exceptions are recorded.
Explanation
Correct Answer
B:Compliant test creations succeed and prohibited configurations are rejected with useful error evidence. This directly tests whether the Custom organization policy on resource attributes goal was met. Custom organization constraints restrict supported resource configurations using explicit evaluated rules.
Incorrect Answers
A:Unapproved external principals cannot be added to controlled IAM policies. This is evidence about Restricting allowed IAM member domains, not direct proof of the required Custom organization policy on resource attributes behavior. The relevant check is that compliant test creations succeed and prohibited configurations are rejected with useful error evidence.
C:Test users complete sign-in through the designated provider and break-glass access remains available. This is evidence about SAML enterprise single sign-on, not direct proof of the required Custom organization policy on resource attributes behavior. The relevant check is that compliant test creations succeed and prohibited configurations are rejected with useful error evidence.
D:Unapproved VMs cannot acquire public IP addresses while approved exceptions are recorded. This is evidence about Restricting external IPv4 addresses through org policy, not direct proof of the required Custom organization policy on resource attributes behavior. The relevant check is that compliant test creations succeed and prohibited configurations are rejected with useful error evidence.
Question 6
A production batch worker needs access to one storage bucket but not project-wide administration. Which practical correction or implementation decision is MOST directly appropriate?
- Review runtime identity bindings and test legitimate and prohibited API calls.
- Put an owner's personal user credentials in every container so all workloads inherit user permissions.
- Review impersonation IAM policy and correlate caller and delegate audit activity.
- Test Kubernetes subject selectors and resulting Secret Manager access from multiple namespaces.
Explanation
Correct Answer
A:Review runtime identity bindings and test legitimate and prohibited API calls. The intended behavior can be checked because each workload can use its intended resource and fails to read another workload's protected data.
Incorrect Answers
B:Put an owner's personal user credentials in every container so all workloads inherit user permissions. This would not correct the Service account per-workload identity design workflow. One broadly shared runtime principal turns a minor service compromise into cross-application access.
C:Review impersonation IAM policy and correlate caller and delegate audit activity. This could help Service account impersonation and token creation, but it does not resolve the Service account per-workload identity design case. Service accounts represent machine identities and should not be shared indiscriminately across trust boundaries.
D:Test Kubernetes subject selectors and resulting Secret Manager access from multiple namespaces. This could help GKE workload identity binding, but it does not resolve the Service account per-workload identity design case. Service accounts represent machine identities and should not be shared indiscriminately across trust boundaries.
Question 7
An automation service maintains a security group for a rotating set of approved analysts. Which approach BEST meets the requirement while respecting the system constraints?
- Compare authoritative HR events, group membership changes and effective IAM grants using audit evidence.
- Configure Google Cloud Directory Sync with scoped mapping rules and validate dry-run changes.
- Use supported Cloud Identity Groups APIs with narrowly scoped permissions and review group change logs.
- Edit a project label and assume it changes Cloud Identity security group membership.
Explanation
Correct Answer
C:Use supported Cloud Identity Groups APIs with narrowly scoped permissions and review group change logs. Managing authorization through groups can simplify changes but group administration itself is privileged. Verification should demonstrate that membership changes match approved source records and unexpected external identities are rejected.
Incorrect Answers
A:Compare authoritative HR events, group membership changes and effective IAM grants using audit evidence. This primarily addresses Group-based provisioning audit trail, not the specific Programmatic group membership governance decision. Here, managing authorization through groups can simplify changes but group administration itself is privileged.
B:Configure Google Cloud Directory Sync with scoped mapping rules and validate dry-run changes. This primarily addresses Cloud Identity directory synchronization, not the specific Programmatic group membership governance decision. Here, managing authorization through groups can simplify changes but group administration itself is privileged.
D:Edit a project label and assume it changes Cloud Identity security group membership. This is an unsafe or ineffective shortcut for Programmatic group membership governance. Unverified group automation may silently grant inherited IAM roles to unintended principals.
Question 8
A new project is created and its logs should flow to the central security bucket without manual per-project setup. An engineer must explain the recommendation for Organization aggregated log sink. Which underlying principle is the MOST accurate?
- Aggregated sinks centralize eligible descendant logs but need correct scope and sink writer identity access.
- Posture findings describe detected conditions and do not equal an enforcement or repair action.
- Object access investigation combines API audit events, identity state and potential downstream egress evidence.
- Guest OS patching remains an operational responsibility despite cloud-managed physical infrastructure.
Explanation
Correct Answer
A:Aggregated sinks centralize eligible descendant logs but need correct scope and sink writer identity access. Therefore the implementation decision is to configure an aggregated log sink with appropriate inclusion filters and destination writer permissions.
Incorrect Answers
B:A log sink lacking destination write permission can silently fail central collection. Posture findings describe detected conditions and do not equal an enforcement or repair action.
C:A log sink lacking destination write permission can silently fail central collection. This describes the separate Cloud Storage audit trail for exfiltration mechanism; the present case instead requires Organization aggregated log sink.
D:A log sink lacking destination write permission can silently fail central collection. This describes the separate VM Manager patch orchestration mechanism; the present case instead requires Organization aggregated log sink.
Question 9
A document retrieval agent reads a user-uploaded file containing instructions to reveal other users' records. Which approach BEST meets the requirement while respecting the system constraints?
- Record trusted artifact provenance, integrity hashes and deployment approvals for model packages.
- Restrict model training service identities and source datasets with IAM and network boundaries.
- Execute arbitrary instructions found in a document because the model describes them as system policy.
- Enforce tool authorization and distinguish untrusted retrieved content from trusted application instructions.
Explanation
Correct Answer
D:Enforce tool authorization and distinguish untrusted retrieved content from trusted application instructions. Prompt-injection defenses require trusted instruction boundaries and downstream authorization, not only content filters. Verification should demonstrate that the agent ignores unauthorized data-exfiltration requests and tool calls remain within caller permissions.
Incorrect Answers
A:Record trusted artifact provenance, integrity hashes and deployment approvals for model packages. This primarily addresses AI model artifact provenance and integrity, not the specific Prompt injection and retrieved data boundary decision. Here, prompt-injection defenses require trusted instruction boundaries and downstream authorization, not only content filters.
B:Restrict model training service identities and source datasets with IAM and network boundaries. This primarily addresses AI training data isolation and least privilege, not the specific Prompt injection and retrieved data boundary decision. Here, prompt-injection defenses require trusted instruction boundaries and downstream authorization, not only content filters.
C:Execute arbitrary instructions found in a document because the model describes them as system policy. This is an unsafe or ineffective shortcut for Prompt injection and retrieved data boundary. Treating retrieved text as a privileged system command can disclose protected data.
Question 10
An old key remains active after engineers migrated to managed runtime identity. Testing exposes this failure: Rotating only a repository secret variable leaves an already leaked key valid. Which remediation should be applied FIRST?
- Validate key disablement through audit logs and test legitimate replacement authentication.
- Create distinct runtime service accounts and bind task-specific roles at the smallest supported scope.
- Remove the public Git commit and assume the exposed key cannot be copied or used.
- Inventory workloads using the identity and disable or suspend the affected service account with a recovery plan.
Explanation
Correct Answer
A:Validate key disablement through audit logs and test legitimate replacement authentication. Confirmation requires seeing that the leaked key can no longer authenticate and supported workloads still operate using their new identity.
Incorrect Answers
B:Create distinct runtime service accounts and bind task-specific roles at the smallest supported scope. It addresses Service account per-workload identity design, not the diagnosed Service account key inventory and exposure response failure. Rotating only a repository secret variable leaves an already leaked key valid.
C:Remove the public Git commit and assume the exposed key cannot be copied or used. This retains the Service account key inventory and exposure response problem. Long-lived service account keys are exportable credentials that require strict inventory and incident response.
D:Inventory workloads using the identity and disable or suspend the affected service account with a recovery plan. It addresses Service account disabling and dependency mapping, not the diagnosed Service account key inventory and exposure response failure. Rotating only a repository secret variable leaves an already leaked key valid.