Question 340An application uses two static routes to the same destination with conflicting priorities and expects the higher numeric priority to win. The team proposes: Adjust route priority based on Google Cloud's lower-number-wins route preference and verify the selected path. Which networking principle explains why this change is appropriate? Pod scheduling depends on alias-IP range capacity, not just available node CPU. Route priority ordering must be interpreted correctly when multiple eligible next hops match. Private service access depends on supported DNS names and reachable endpoints, not arbitrary address rewrites. Centralized network scaling requires visibility into effective propagation and filter decisions. ExplanationCorrect Answer B:Route priority ordering must be interpreted correctly when multiple eligible next hops match. This is the control that makes the correction effective. Adjust route priority based on Google Cloud's lower-number-wins route preference and verify the selected path. A working result shows: A route listing and packet trace show traffic using the intended preferred path. Incorrect Answers A:Pod scheduling depends on alias-IP range capacity, not just available node CPU. This does not establish the proposed intervention. The governing principle here is: Route priority ordering must be interpreted correctly when multiple eligible next hops match. Apply it by: Adjust route priority based on Google Cloud's lower-number-wins route preference and verify the selected path. C:Private service access depends on supported DNS names and reachable endpoints, not arbitrary address rewrites. This does not establish the proposed intervention. The governing principle here is: Route priority ordering must be interpreted correctly when multiple eligible next hops match. Apply it by: Adjust route priority based on Google Cloud's lower-number-wins route preference and verify the selected path. D:Centralized network scaling requires visibility into effective propagation and filter decisions. This does not establish the proposed intervention. The governing principle here is: Route priority ordering must be interpreted correctly when multiple eligible next hops match. Apply it by: Adjust route priority based on Google Cloud's lower-number-wins route preference and verify the selected path.
Question 341A cloud migration creates production and disaster-recovery networks with identical internal CIDRs but later attempts to connect them through VPC peering. Which Google Cloud networking change should the engineer make first? Inspect NCC propagated routes, spoke association and range filters before changing firewall rules. Update the authorized network list to the current verified operator source ranges. Allocate nonoverlapping CIDRs before establishing the unsupported overlapping peering topology. Verify supported bidirectional routing and stateful appliance insertion for the targeted flows. ExplanationCorrect Answer C:Allocate nonoverlapping CIDRs before establishing the unsupported overlapping peering topology. VPC peering requires compatible unique subnet routes between networks. Its operational proof is: Peering can be established and reachability tests select unambiguous destination ranges. The avoidable impact is: Overlapping routes can block peering or isolate critical recovery services. Incorrect Answers A:Spoke health and destination-specific connectivity are separate states. That is relevant to a different intervention. For the stated situation, the required action is: Allocate nonoverlapping CIDRs before establishing the unsupported overlapping peering topology. The governing reason is: VPC peering requires compatible unique subnet routes between networks. B:Allowlisting access to cluster control planes must track changes to administrator network locations. That is relevant to a different intervention. For the stated situation, the required action is: Allocate nonoverlapping CIDRs before establishing the unsupported overlapping peering topology. The governing reason is: VPC peering requires compatible unique subnet routes between networks. D:Stateful network inspection requires coherent paths for request and response traffic. That is relevant to a different intervention. For the stated situation, the required action is: Allocate nonoverlapping CIDRs before establishing the unsupported overlapping peering topology. The governing reason is: VPC peering requires compatible unique subnet routes between networks.
Question 342A service works on one VM but another VM in the same subnet cannot connect and operators blame the Cloud NAT gateway. After this change — Compare effective firewall policies, tags and service-account identities for both VMs. — which observation would best demonstrate that it worked? The expected VM matches the approved rule and the denied VM has a traceable policy verdict. The required prefix appears in receiving effective routes and test traffic reaches it. A test session interruption triggers an actionable alert for the correct BGP peer. The correct denied flow appears under its matched rule rather than an inferred topology edge. ExplanationCorrect Answer A:The expected VM matches the approved rule and the denied VM has a traceable policy verdict. The relevant control is operating as intended. Same-subnet placement does not imply equivalent firewall targeting or identity policy. An unresolved issue would mean: Misdiagnosing identity-scoped firewall rules can lead to broad unsafe exceptions. Incorrect Answers B:A live BGP session is necessary but not sufficient for learning a specific destination route. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: The expected VM matches the approved rule and the denied VM has a traceable policy verdict. C:Routing session availability needs timely control-plane monitoring beyond periodic manual inspection. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: The expected VM matches the approved rule and the denied VM has a traceable policy verdict. D:Topology visualization and packet-level firewall diagnostics answer different questions. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: The expected VM matches the approved rule and the denied VM has a traceable policy verdict.
Question 343A multinational retailer chooses Standard Tier for an externally facing application that needs globally optimized anycast entry points. Which Google Cloud networking change should the engineer make first? Plan Shared VPC or Network Connectivity Center hub-and-spoke connectivity with supported route exchange. Define the right mix of standalone and Shared VPC networks with explicit trust boundaries. Distribute critical workloads over adequately sized node pools and planned zones. Select Premium Tier and a compatible global external load balancer for the required worldwide entry behavior. ExplanationCorrect Answer D:Select Premium Tier and a compatible global external load balancer for the required worldwide entry behavior. Network tier determines eligible global edge features and traffic paths, independently of application autoscaling. The evidence to require before closing the incident is: A probe from three continents reaches the approved global frontend with consistent backend health. Incorrect Answers A:Transit across VPC peerings is not automatically provided by multiple bilateral connections. That is relevant to a different intervention. For the stated situation, the required action is: Select Premium Tier and a compatible global external load balancer for the required worldwide entry behavior. The governing reason is: Network tier determines eligible global edge features and traffic paths, independently of application autoscaling. B:Shared connectivity and administrative isolation are separate design requirements, not a single default choice. That is relevant to a different intervention. For the stated situation, the required action is: Select Premium Tier and a compatible global external load balancer for the required worldwide entry behavior. The governing reason is: Network tier determines eligible global edge features and traffic paths, independently of application autoscaling. C:GKE networking availability also depends on node and scheduling failure domains. That is relevant to a different intervention. For the stated situation, the required action is: Select Premium Tier and a compatible global external load balancer for the required worldwide entry behavior. The governing reason is: Network tier determines eligible global edge features and traffic paths, independently of application autoscaling.
Question 344A company stores Cloud DNS query logs from every application indefinitely without reviewing sensitive-name retention. The team proposes: Apply approved DNS logging scope, redaction access and retention controls. Which networking principle explains why this change is appropriate? Useful DNS diagnostics can also expose internal application and user activity metadata. A visible logical connection is not evidence that every required destination prefix is exchanged. An application response can succeed even when the traffic uses an expensive or unsafe route. Health checks do not necessarily prove connection continuity during planned backend changes. ExplanationCorrect Answer A:Useful DNS diagnostics can also expose internal application and user activity metadata. That is why the proposed change is suitable: Apply approved DNS logging scope, redaction access and retention controls. A direct check is: A test analyst sees authorized query metrics but cannot retrieve unrelated sensitive historical names. Incorrect Answers B:A visible logical connection is not evidence that every required destination prefix is exchanged. This does not establish the proposed intervention. The governing principle here is: Useful DNS diagnostics can also expose internal application and user activity metadata. Apply it by: Apply approved DNS logging scope, redaction access and retention controls. C:An application response can succeed even when the traffic uses an expensive or unsafe route. This does not establish the proposed intervention. The governing principle here is: Useful DNS diagnostics can also expose internal application and user activity metadata. Apply it by: Apply approved DNS logging scope, redaction access and retention controls. D:Health checks do not necessarily prove connection continuity during planned backend changes. This does not establish the proposed intervention. The governing principle here is: Useful DNS diagnostics can also expose internal application and user activity metadata. Apply it by: Apply approved DNS logging scope, redaction access and retention controls.
Question 345A security operations team enables VPC Flow Logs on the wrong subnet and concludes a missing payment flow never existed. After this change — Enable the appropriate subnet flow logging and correlate it with firewall and service logs. — which observation would best demonstrate that it worked? A controlled payment attempt generates a matching sampled flow record at the expected subnet. A synthetic route loss creates the expected spoke alert and affected destination is identified. A route trace confirms the permitted intended attachment and an outage test confirms fallback. A controlled request trace identifies the slow path component under representative traffic. ExplanationCorrect Answer A:A controlled payment attempt generates a matching sampled flow record at the expected subnet. That test checks the affected service boundary directly. Enable the appropriate subnet flow logging and correlate it with firewall and service logs. Without it, missing telemetry can produce false conclusions during an incident investigation. Incorrect Answers B:Topology monitoring must track the actual connectivity fabric after architecture migration. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: A controlled payment attempt generates a matching sampled flow record at the expected subnet. C:An application response can succeed even when the traffic uses an expensive or unsafe route. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: A controlled payment attempt generates a matching sampled flow record at the expected subnet. D:Observed network latency can include routing, load-balancing and endpoint contributions. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: A controlled payment attempt generates a matching sampled flow record at the expected subnet.
Question 346A checkout service needs to drain existing connections during a backend upgrade but simply deletes healthy instances immediately. What is the most direct consequence of leaving the problem uncorrected? An internal-only frontend can make a public product inaccessible. Abrupt backend removal can drop payment transactions mid-flight. Hidden private zones can make otherwise routable services unreachable. Unauthenticated asset URLs can make licensed content broadly available. ExplanationCorrect Answer B:Abrupt backend removal can drop payment transactions mid-flight. This is the most immediate service or security consequence. Connection draining separates new connection admission from completing existing client transactions. Test the correction by verifying: Existing sessions finish while new requests move to healthy replacement backends. Incorrect Answers A:A load balancer's scope and protocol behavior determine where users can reach application entry points. That does not describe the principal exposure from the stated failure. The direct avoidable risk is: Abrupt backend removal can drop payment transactions mid-flight. C:Private DNS visibility depends on authorized network association, not merely organization ownership. That does not describe the principal exposure from the stated failure. The direct avoidable risk is: Abrupt backend removal can drop payment transactions mid-flight. D:CDN caching alone does not authenticate requests for restricted media. That does not describe the principal exposure from the stated failure. The direct avoidable risk is: Abrupt backend removal can drop payment transactions mid-flight.
Question 347A cloud operator configures out-of-band Network Security Integration but assumes an inline appliance health check proves the separate threat telemetry pipeline works. What is the most direct consequence of leaving the problem uncorrected? A single firewall VM can become a network-wide inspection outage point. Regional-only controls can leave other regions unprotected. Silent telemetry failure can leave an apparently healthy appliance without detection coverage. Relying on the wrong layer can leave infrastructure vulnerable during large attacks. ExplanationCorrect Answer C:Silent telemetry failure can leave an apparently healthy appliance without detection coverage. The described failure makes this outcome plausible. Validate the authorized out-of-band event forwarding and its collector processing independently. The acceptance check is: A controlled suspicious flow creates an observable event at the approved integration endpoint. Incorrect Answers A:Stateful virtual appliance insertion requires both routing symmetry and resilient backend capacity. That does not describe the principal exposure from the stated failure. The direct avoidable risk is: Silent telemetry failure can leave an apparently healthy appliance without detection coverage. B:Policy scope determines which regions and networks inherit an enforcement requirement. That does not describe the principal exposure from the stated failure. The direct avoidable risk is: Silent telemetry failure can leave an apparently healthy appliance without detection coverage. D:Volumetric and application-layer attacks require correctly placed controls at different layers. That does not describe the principal exposure from the stated failure. The direct avoidable risk is: Silent telemetry failure can leave an apparently healthy appliance without detection coverage.
Question 348A public API receives automated credential-stuffing attempts but its Cloud Armor policy does not distinguish bot behavior from normal clients. The team proposes: Evaluate supported bot-management controls with approved challenge and access policy. Which networking principle explains why this change is appropriate? Cloud network microsegmentation requires target selectors that actually represent the running workload. Third-party allowlists need predictable source addresses after NAT translation. Out-of-band security integration is a detection workflow distinct from the traffic-forwarding data plane. Automated-abuse mitigation must balance bot signals against legitimate user authentication flows. ExplanationCorrect Answer D:Automated-abuse mitigation must balance bot signals against legitimate user authentication flows. This is the control that makes the correction effective. Evaluate supported bot-management controls with approved challenge and access policy. A working result shows: A synthetic bot-pattern test triggers the intended mitigation without blocking normal users. Incorrect Answers A:Cloud network microsegmentation requires target selectors that actually represent the running workload. This does not establish the proposed intervention. The governing principle here is: Automated-abuse mitigation must balance bot signals against legitimate user authentication flows. Apply it by: Evaluate supported bot-management controls with approved challenge and access policy. B:Third-party allowlists need predictable source addresses after NAT translation. This does not establish the proposed intervention. The governing principle here is: Automated-abuse mitigation must balance bot signals against legitimate user authentication flows. Apply it by: Evaluate supported bot-management controls with approved challenge and access policy. C:Out-of-band security integration is a detection workflow distinct from the traffic-forwarding data plane. This does not establish the proposed intervention. The governing principle here is: Automated-abuse mitigation must balance bot signals against legitimate user authentication flows. Apply it by: Evaluate supported bot-management controls with approved challenge and access policy.
Question 349A video platform places its only ingress load balancer in one region despite a contractual regional-outage requirement. Which Google Cloud networking change should the engineer make first? Define directional forwarding and DNS peering paths that avoid cyclic resolution. Choose a compatible Pod range and update the hybrid IPAM registry before cluster build. Design geographically redundant backends and the appropriate cross-region frontend failover strategy. Design supported policy-based routing or an internal load-balancer next hop for a resilient appliance pool. ExplanationCorrect Answer C:Design geographically redundant backends and the appropriate cross-region frontend failover strategy. Availability must consider an entire regional failure domain, not merely redundant instances in one zone. Its operational proof is: Disabling a primary-region backend leaves approved requests served by a healthy secondary region. The avoidable impact is: A regional incident could interrupt every subscriber session. Incorrect Answers A:Hybrid DNS forwarding requires explicit authority and one-way recursion boundaries. That is relevant to a different intervention. For the stated situation, the required action is: Design geographically redundant backends and the appropriate cross-region frontend failover strategy. The governing reason is: Availability must consider an entire regional failure domain, not merely redundant instances in one zone. B:Pod ranges participate in real network reachability and must not conflict with hybrid destination prefixes. That is relevant to a different intervention. For the stated situation, the required action is: Design geographically redundant backends and the appropriate cross-region frontend failover strategy. The governing reason is: Availability must consider an entire regional failure domain, not merely redundant instances in one zone. D:Inspection insertion must maintain both routing intent and a healthy appliance failure domain. That is relevant to a different intervention. For the stated situation, the required action is: Design geographically redundant backends and the appropriate cross-region frontend failover strategy. The governing reason is: Availability must consider an entire regional failure domain, not merely redundant instances in one zone.
Question 350A Kubernetes platform deploys external-dns against Cloud DNS but grants it broad organization-wide DNS administration. After this change — Restrict its service identity to the specific zone and record-change permissions required. — which observation would best demonstrate that it worked? A datacenter client resolves the approved cloud-private zone without recursive loops. An approved cross-region client reaches the service without opening an internet path. A test media URL is served with expected cache headers and CDN hit telemetry. The controller creates its authorized service record but cannot alter unrelated zones. ExplanationCorrect Answer D:The controller creates its authorized service record but cannot alter unrelated zones. The relevant control is operating as intended. Automated DNS record management should operate at the least practical privilege. An unresolved issue would mean: Overbroad DNS automation privileges can redirect unrelated production names. Incorrect Answers A:Hybrid private DNS requires a reachable inbound resolver path and correct forwarding rules. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: The controller creates its authorized service record but cannot alter unrelated zones. B:Regional internal load-balancer reachability is not automatically universal across every network location. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: The controller creates its authorized service record but cannot alter unrelated zones. C:Cloud CDN origin support depends on an eligible backend integration with the serving load balancer. This relates to a separate check and does not verify the specified repair. The relevant acceptance evidence is: The controller creates its authorized service record but cannot alter unrelated zones.