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 manage a VPC-native GKE Standard cluster in a Shared VPC running external web services and sensitive payment-processing workloads. You must transition to a heterogeneous node pool design meeting these requirements:
LoadBalancer Service (externalTrafficPolicy: Local) to the private pool without downtime, and freeze automatic upgrades only on the private pool.Which architecture and operational procedure should you implement?
This approach pairs --enable-private-nodes and node-restriction.kubernetes.io/ scheduling controls with automatic Private Google Access, Cloud NAT, an immediate cordon and drain workflow, and a node pool maintenance exclusion.
LoadBalancer Service uses externalTrafficPolicy: Local, load balancer health checks only pass on nodes currently hosting ready backend Pods. Immediately cordoning and draining the existing pool before payment Pods are running and ready on the private pool removes all healthy backends, blackholing traffic until rescheduling completes.externalTrafficPolicy: Local preserves client source IP addresses by routing traffic strictly to nodes with local Pods, which makes pre-deployment mandatory before draining old nodes.The cluster resided in a standalone VPC (where Private Google Access is enabled automatically) and the Service used externalTrafficPolicy: Cluster, permitting cross-node forwarding during Pod rescheduling.
When migrating services configured with externalTrafficPolicy: Local, always deploy replicas onto the destination node pool and verify readiness before cordoning the source node pool, or temporarily switch to externalTrafficPolicy: Cluster during the transition.
This approach creates a public node pool protected by VPC firewall rules, applies node-restriction.kubernetes.io/ labels and taints, enables Private Google Access without Cloud NAT, performs an overlap-safe Pod migration, and sets a node pool maintenance exclusion.
--enable-private-nodes assigns public IP addresses to every node in the payment pool. Relying solely on VPC firewall rules to block inbound traffic violates the requirement to isolate sensitive workloads on private nodes without external IP addresses.--enable-private-nodes is a node-pool-level setting that removes external IPs from node network interfaces, eliminating direct internet routing at the addressing layer rather than relying solely on firewall filtering.private.googleapis.com or restricted.googleapis.com), not general internet egress.External IP addresses on payment nodes were permitted by compliance policies, and workloads only needed to reach Google Cloud APIs rather than third-party internet endpoints.
In a Shared VPC, combining --enable-private-nodes with explicitly enabled Private Google Access and a Cloud NAT gateway configured for the cluster subnet ranges provides complete, secure outbound connectivity without public node IPs.
A heterogeneous GKE node pool architecture runs public and private node pools side by side in a VPC-native cluster by applying --enable-private-nodes per node pool, combined with hardened Kubernetes scheduling, Shared VPC private egress, overlap-safe service migration, and scoped maintenance policies.
--enable-private-nodes removes external IP addresses from those nodes while keeping the web pool public. Applying node taints with node-restriction.kubernetes.io/ labels—and matching tolerations and node affinity on payment Pods—ensures bidirectional isolation, while the NodeRestriction plugin prevents compromised kubelet credentials from modifying those isolation labels.externalTrafficPolicy: Local health checks passing on the new nodes during cutover. A node pool maintenance exclusion freezes automatic upgrades strictly on the private pool while allowing the public pool to upgrade automatically.This design addresses every security, Shared VPC routing, load balancer backend, and lifecycle constraint using native GKE capabilities without requiring separate clusters.
This configuration uses --enable-private-nodes, standard Kubernetes node labels with node affinity, explicit Private Google Access and Cloud NAT, an overlap-safe Pod migration, and a cluster-level maintenance exclusion.
node-restriction.kubernetes.io/ prefix can be modified by a node's kubelet using its own node credentials. A compromised node could add tier=payment to itself and attract sensitive payment Pods.No minor or node upgrades freezes automatic upgrades across the entire cluster, blocking automated upgrades on the public node pool as well.kubelet processes from setting or modifying labels prefixed with node-restriction.kubernetes.io/, protecting isolation boundaries.You wanted to freeze upgrades across all node pools in the cluster simultaneously and did not need protection against kubelet label modification or non-sensitive Pods co-scheduling on the private pool.
In heterogeneous clusters, pair node pool maintenance exclusions with both taints and node-restriction.kubernetes.io/ labels to enforce strict, tamper-resistant separation between public and private workloads.
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