Prepare and test your skills

Prepare and test your skills

Free sample question with answers and detailed explanation below. The Professional Cloud Network Engineer practice bank has an additional 2,437 practice questions.Try 10 free realistic questions: instant score, no credit card, and no account required.
Your organization runs a VPC-native private GKE cluster with GKE Dataplane V2 enabled, hosting multi-tenant microservices across separate namespaces. You are migrating from legacy VPC firewall rules to a Cloud NGFW global network firewall policy (Standard tier) and must enforce defense-in-depth security across node and Pod boundaries:
api.partner.example.com on TCP 443) exclusively through a dedicated gateway node pool, and preserve cluster control-plane and inter-node communications.type: LoadBalancer with GKE subsetting) accessible only from 10.20.0.0/16, while ensuring Google Cloud health checks succeed.Which configuration should you implement?
100 egress deny-all rule (0.0.0.0/0) and a priority 50 egress allow rule to api.partner.example.com:443 targeted at the gateway node pool, relying on GKE's automatically created firewall rules to maintain control-plane and inter-node connectivity. Set spec.loadBalancerSourceRanges to 10.20.0.0/16 on the internal Service and deploy namespace NetworkPolicy resources that allow DNS (kube-system:53) and gateway egress.This approach pairs a high-priority (priority 100) Cloud NGFW egress deny-all rule with a priority 50 FQDN allow rule for the gateway node pool, assuming that GKE's automatically provisioned firewall rules will keep control-plane, Konnectivity, and node-to-Pod traffic operational.
1000 or 999) and depends on the VPC's implied lowest-priority egress allow rule for outbound traffic. Introducing an explicit egress deny-all rule at priority 100 without higher-priority egress allow rules immediately blocks nodes from reaching the Kubernetes control plane (TCP 443), the Konnectivity proxy (TCP 8132), and other nodes or Pods across all protocols and ports.1000, a Cloud NGFW policy rule at priority 100 is evaluated earlier and would drop essential cluster traffic before lower-priority rules are reached.spec.loadBalancerSourceRanges still leaves GKE creating legacy VPC ingress rules rather than managing access cleanly through Cloud NGFW policies.65535) or explicitly higher-priority allow rules for TCP 443, TCP 8132, and cluster node/Pod CIDRs.The egress deny-all rule were placed at a low priority (65535) and paired with higher-priority Cloud NGFW egress allow rules for the control plane (443, 8132) and node/Pod CIDRs, alongside Cloud NGFW ingress rules for the internal LoadBalancer Service and health checks.
65535 egress deny-all rule (0.0.0.0/0), higher-priority egress allow rules for control-plane/Konnectivity (443, 8132) and node/Pod CIDRs (all protocols), and a gateway-node-targeted egress allow rule for api.partner.example.com:443. Disable automatic LoadBalancer firewall rules (or set loadBalancerSourceRanges to 10.20.0.0/16), allow 10.20.0.0/16 and health check CIDRs (35.191.0.0/16, 130.211.0.0/22) via Cloud NGFW ingress rules, and configure namespace NetworkPolicy resources denying cross-namespace traffic while allowing DNS (kube-system:53) and gateway egress.This configuration establishes a defense-in-depth network security model by combining Cloud NGFW global network firewall policies at the VPC/node boundary with Kubernetes NetworkPolicy resources (enforced by GKE Dataplane V2) at the Pod and namespace level.
0.0.0.0/0) rule at priority 65535 blocks unauthorized internet access while higher-priority egress allow rules explicitly preserve node-to-control-plane traffic (TCP 443 and 8132 for the Konnectivity proxy) and inter-node/Pod CIDR communication (all protocols and ports). Targeting the FQDN object rule (api.partner.example.com:443) solely to the gateway node pool's service account or secure tag ensures application nodes cannot bypass egress proxies.spec.loadBalancerSourceRanges to 10.20.0.0/16) prevents open 0.0.0.0/0 access, while Cloud NGFW ingress allow rules explicitly permit 10.20.0.0/16 and Google Cloud health check probes (35.191.0.0/16 and 130.211.0.0/22).NetworkPolicy rules enforce default-deny across namespaces while allowing essential DNS queries to kube-system on port 53, traffic to gateway Pods, and authorized Service ports.Neither Cloud NGFW nor Kubernetes NetworkPolicy alone secures every layer. Combining node-level Cloud NGFW policies with Pod-level NetworkPolicy and restricted Service definitions protects against both external threats and lateral movement.
65535 egress deny-all rule, higher-priority egress allow rules for control-plane (443, 8132) and gateway-node access to api.partner.example.com:443, and east-west Cloud NGFW rules using Pod IP CIDRs and secure tags to block cross-namespace Pod-to-Pod traffic. Configure spec.loadBalancerSourceRanges to 10.20.0.0/16 and allow health check ranges via Cloud NGFW ingress rules, omitting Kubernetes NetworkPolicy objects.This design correctly structures the Cloud NGFW egress rules for the control plane and gateway nodes, but attempts to replace Kubernetes NetworkPolicy resources entirely with Cloud NGFW east-west rules based on Pod IP ranges and secure tags.
NetworkPolicy.65535 deny-all rule can disrupt cross-node Pod communication.NetworkPolicy scope: Enforces Layer 3 and Layer 4 rules using Pod labels and namespace selectors, automatically updating enforcement as Pods are scheduled.Namespace-scoped Kubernetes NetworkPolicy objects were deployed to handle east-west Pod isolation (while permitting kube-system DNS on port 53) and Cloud NGFW rules were reserved for node-level ingress, egress, and inter-node/Pod CIDR connectivity.
10.20.0.0/16 while leaving spec.loadBalancerSourceRanges unset on the Kubernetes Service. Enforce cluster-wide outbound internet restrictions solely with Kubernetes FQDNNetworkPolicy and default-deny NetworkPolicy resources (allowing kube-system:53 and api.partner.example.com:443), without adding a Cloud NGFW egress deny rule.This option attempts to attach Cloud NGFW ingress rules to managed Envoy proxies for an internal passthrough Network Load Balancer, leaves spec.loadBalancerSourceRanges undefined on the Service, and relies exclusively on Kubernetes NetworkPolicy and FQDNNetworkPolicy for egress control.
spec.loadBalancerSourceRanges is omitted on a LoadBalancer Service and automatic firewall creation remains enabled, GKE creates an ingress allow rule permitting 0.0.0.0/0 on the Service ports, exposing it beyond 10.20.0.0/16.loadBalancerSourceRanges.The Service used an internal Application Load Balancer or internal proxy Network Load Balancer (for Envoy proxy targeting) or targeted GKE node VMs directly with loadBalancerSourceRanges restricted, and a Cloud NGFW egress deny-all policy were configured alongside Kubernetes network policies.
Answer them yourself and get your score the moment you finish. No account needed.
Full access includes
Keep the momentum going with these hand-picked practice scenarios
Last updated Validated