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.
You are planning IP addressing for multiple large GKE Standard VPC-native clusters (version 1.30) in a Shared VPC connected to an on-premises data center via Cloud Interconnect and to another VPC via VPC Network Peering. Your environment has the following constraints and requirements:
/24 RFC 1918 blocks remain available, and the organization owns no unused public IPv4 space (100.64.0.0/10 is unused).Which addressing strategy should you implement?
100.64.0.0/10 (RFC 6598) ranges for Pod secondary ranges, use the GKE-managed 34.118.224.0/20 range for Services, and configure the IP masquerade agent (nonMasqueradeCIDRs) to exclude the cluster and connected internal CIDRs from SNAT.100.64.0.0/10) and GKE-managed Service range strategy isRFC 6598 (100.64.0.0/10) is the Shared Address Space originally designated for Carrier-Grade NAT (CGN), providing over 4 million non-RFC 1918 private IPv4 addresses. Pairing RFC 1918 /24 primary ranges for nodes (and internal load balancers) with 100.64.0.0/10 secondary Pod ranges, the GKE-managed 34.118.224.0/20 range for ClusterIP Services, and customized IP masquerade agent (ip-masq-agent) rules resolves RFC 1918 exhaustion while maintaining a fully integrated GKE networking model.
/24 RFC 1918 ranges are consumed by node subnets, while millions of Pod IPs are carved from 100.64.0.0/10 and Services (ClusterIP) consume zero customer subnet space by using Google's internal 34.118.224.0/20 range.240.0.0.0/4), 100.64.0.0/10 is widely accepted by Windows operating systems and on-premises enterprise routers. Unlike Privately Used Public IPs (PUPIs), extended private addresses in 100.64.0.0/10 are automatically advertised over VPC Network Peering and can be advertised over Cloud Interconnect.nonMasqueradeCIDRs in ip-masq-agent to include the Pod, Service, on-premises, and peered VPC ranges prevents SNAT for internal and hybrid traffic while still allowing SNAT for internet egress.100.64.0.0/10 nor 34.118.224.0/20 overlaps with publicly routable internet destinations.ClusterIP Services (available by default in GKE Standard 1.29+).Among non-RFC 1918 options, 100.64.0.0/10 avoids the hardware/OS blocking of Class E (240.0.0.0/4) and the internet-routability conflicts and VPC peering restrictions of unowned PUPI blocks, making it the recommended non-RFC 1918 choice for hybrid and peered GKE topologies.
34.118.224.0/20 range for Services, and configure the IP masquerade agent to SNAT outbound Pod traffic to the node IP addresses.Reusing RFC 1918 ranges already allocated elsewhere in an enterprise network while enabling SNAT at the node boundary attempts to mimic an island-mode Kubernetes network model inside GKE. In this pattern, Pod IPs are hidden behind unique node IP addresses for all traffic leaving the cluster.
ip-masq-agent to SNAT outbound Pod traffic translates every Pod's source IP address to the underlying node's primary IP address. External receivers—including on-premises Windows hosts, hardware appliances, and peered VPC workloads—cannot see individual Pod source IPs in application logs or apply Pod-granular firewall rules.34.118.224.0/20 range for Services, and disable default SNAT.PUPI addressing allows a VPC-native GKE cluster to use external public IPv4 ranges (that are not owned by Google) privately as internal subnet or secondary Pod CIDR ranges. While Google Cloud does not require you to own those public IPs to configure them internally, doing so carries major routing and reachability trade-offs.
100.64.0.0/10 ranges, PUPI routes are not automatically imported across VPC Network Peering connections by default; peers must explicitly enable --import-subnet-routes-with-public-ip. Additionally, advertising third-party public IP blocks over Cloud Interconnect requires strict route-advertisement filtering on-premises to prevent internet route leaks.--disable-default-snat or ip-masq-agent customization to prevent Pod-to-Pod SNAT.240.0.0.0/4 (Class E / RFC 5735) ranges for Pod secondary ranges, use the GKE-managed 34.118.224.0/20 range for Services, and disable default SNAT so Pod source IPs are preserved.240.0.0.0/4) addressing isClass E (defined in RFC 5735 / RFC 1112) is the 240.0.0.0/4 IPv4 block originally reserved for future use and experimental networking. Google Cloud permits using 240.0.0.0/4 as a non-RFC 1918 internal range for VPC subnets and GKE Pod secondary ranges, providing over 268 million private IPv4 addresses without conflicting with public internet routing.
240.0.0.0/4 is blocked or dropped by Windows hosts and many legacy on-premises routers, firewalls, and hardware appliances that treat Class E space as invalid or reserved multicast space./4 prefix (over 268 million addresses), which makes it attractive for extreme Pod scale when no other private space is available.100.64.0.0/10 (RFC 6598) were already exhausted elsewhere in the enterprise network, and Pods did not need direct non-NAT connectivity to Windows hosts or incompatible on-premises hardware (allowing SNAT to RFC 1918 node IPs for external/on-premises destinations).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