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 GKE Standard cluster connected to an on-premises network (192.168.0.0/16) via Dedicated Interconnect and a Cloud Router configured with custom route advertisements. Existing network ranges include:
10.10.0.0/2010.20.0.0/1610.30.0.0/24An ip-masq-agent ConfigMap currently lists 10.10.0.0/20, 10.20.0.0/16, 10.30.0.0/24, and 192.168.0.0/16 in nonMasqueradeCIDRs. You deploy a new node pool using an additional Pod range (172.24.0.0/16).
Which configuration changes ensure seamless bidirectional connectivity with preserved Pod IPs across both Pod ranges, VPC endpoints, and on-premises while still masquerading internet egress?
172.24.0.0/16 to nonMasqueradeCIDRs in the ip-masq-agent ConfigMap, update VPC firewall policies for 172.24.0.0/16, and configure a Cloud NAT gateway for the 172.24.0.0/16 secondary subnet range instead of modifying Cloud Router custom advertisements.Cloud NAT (Public NAT) is a regional, VPC-level gateway service that translates internal VPC IP addresses—such as GKE node primary ranges or Pod secondary ranges—into static public IP addresses so instances without external IPs can reach public internet endpoints.
all_subnets, failing to add 172.24.0.0/16 means on-premises routers never learn a BGP route to the new Pod range.ip-masq-agent preserves the 172.24.0.0/16 source IP toward 192.168.0.0/16, on-premises routers will either drop return packets due to missing routes or block on-premises clients from initiating connections to the new Pods.nonMasqueradeCIDRs is set to 0.0.0.0/0).The objective were to allow Pods in 172.24.0.0/16 with SNAT disabled (0.0.0.0/0 in nonMasqueradeCIDRs) to reach public internet services rather than establishing bidirectional connectivity with an on-premises network.
172.24.0.0/16 and 192.168.0.0/16 from nonMasqueradeCIDRs in the ip-masq-agent ConfigMap so traffic from the new Pods is masqueraded to node IPs, and exclude 172.24.0.0/16 from Cloud Router custom advertised IP ranges.Omitting destination ranges from nonMasqueradeCIDRs and excluding Pod ranges from Cloud Router BGP advertisements implements an island-mode emulation pattern that hides Pod IP addresses from external or on-premises networks by translating outbound Pod traffic to the underlying node primary IP address.
172.24.0.0/16 is withheld from BGP advertisements and masked behind node IPs via SNAT, on-premises workloads cannot initiate direct connections to Pod IPs in 172.24.0.0/16, violating the bidirectional connectivity requirement.192.168.0.0/16 forces all GKE Pods—including the original 10.20.0.0/16 range—to lose their source Pod IP identity when communicating with on-premises servers.172.24.0.0/16 from nonMasqueradeCIDRs causes traffic between 10.20.0.0/16 Pods and 172.24.0.0/16 Pods to be SNATed to node IPs, disrupting Pod-level firewalls and cluster network policies.10.10.0.0/20) rather than individual Pod identities.240.0.0.0/4) Pod ranges.The additional Pod range used Class E or overlapping IP addresses that had to be hidden from on-premises routers, and only outbound Pod connections or internal load balancer ingress were required.
172.24.0.0/16 to nonMasqueradeCIDRs in the ip-masq-agent ConfigMap, add 172.24.0.0/16 to the Cloud Router custom advertised IP ranges, and update VPC firewall policies and on-premises firewalls to allow bidirectional traffic for 172.24.0.0/16.In a VPC-native GKE cluster with additional Pod CIDR ranges, seamless bidirectional connectivity requires synchronizing three layers of the network control plane: node-level SNAT rules (ip-masq-agent), BGP route propagation (Cloud Router custom advertisements), and access control policies (VPC firewall policies and on-premises firewalls).
ip-masq-agent ConfigMap defines nonMasqueradeCIDRs, it completely replaces GKE's default non-masquerade list. Adding 172.24.0.0/16 ensures that packets sent from original Pods (10.20.0.0/16) or nodes to the new Pod range retain their original source IPs instead of being translated to node IPs.0.0.0.0/0 is not in nonMasqueradeCIDRs, outbound traffic destined for public internet addresses still undergoes Source Network Address Translation (SNAT) to the node's primary IP address before reaching Cloud NAT.172.24.0.0/16 to --set-advertisement-ranges propagates the route to on-premises routers so return and inbound traffic can reach the new Pods.172.24.0.0/16 allows direct Pod-to-Pod, Pod-to-VPC-endpoint (10.30.0.0/24), and Pod-to-on-premises flows.Unlike approaches that mask Pod IPs via SNAT or rely on internet-only gateways like Cloud NAT, this configuration aligns the dataplane, BGP control plane, and firewall enforcement so the additional Pod range behaves identically to the original Pod range across cloud and hybrid environments.
172.24.0.0/16 to the Cloud Router custom advertised IP ranges and update VPC firewall policies for 172.24.0.0/16, while leaving the ip-masq-agent ConfigMap unchanged because 172.24.0.0/16 is covered by GKE's default RFC 1918 non-masquerade list.ip-masq-agent ConfigMap override doesIn GKE Standard clusters, the ip-masq-agent DaemonSet manages iptables SNAT rules on each Linux node. While GKE maintains a built-in list of default non-masquerade destinations that includes all RFC 1918 ranges (10.0.0.0/8, 172.16.0.0/12, 192.168.0.0/16), those defaults only apply when no custom nonMasqueradeCIDRs list is configured.
nonMasqueradeCIDRs list in the ip-masq-agent ConfigMap entirely replaces—rather than appends to—the default non-masquerade destinations.10.10.0.0/20, 10.20.0.0/16, 10.30.0.0/24, and 192.168.0.0/16, the new range 172.24.0.0/16 is excluded. Any traffic sent from Pods in 10.20.0.0/16 to Pods in 172.24.0.0/16 will be masqueraded (SNATed) to the source node's IP address.ip-masq-agent ConfigMap is evaluated as an exact allowlist for bypassing SNAT; unlisted ranges always fall back to node-IP masquerading.nonMasqueradeCIDRs fails to cover all assigned Pod CIDR ranges.172.16.0.0/12 default is a common misconception whenever a custom ConfigMap already exists in kube-system.No custom ip-masq-agent ConfigMap existed in the cluster and --disable-default-snat were not set, in which case GKE would automatically preserve Pod IPs for all RFC 1918 destinations including 172.24.0.0/16.
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