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 is deploying a Standard VPC-native GKE cluster in us-central1 for regulated workloads. To balance strict network isolation with operational automation, the cluster must meet the following requirements:
10.50.0.0/24) located in us-east1 within the same VPC network.9443, exposed internally by a Kubernetes Service on port 443.Which configuration should you implement?
--enable-private-nodes, --enable-dns-access (with IP-based endpoints disabled), --enable-master-authorized-networks, and --master-authorized-networks=10.50.0.0/24. Create a Kubernetes NetworkPolicy allowing ingress on TCP port 9443 from masterIpv4CidrBlock to the webhook Pods.This configuration enables private nodes, activates the DNS-based control plane endpoint while disabling all IP-based endpoints, passes --enable-master-authorized-networks with 10.50.0.0/24, and defines a Kubernetes NetworkPolicy for TCP port 9443.
--master-authorized-networks.NetworkPolicy cannot open VPC firewalls: NetworkPolicy operates strictly at the Pod level inside the cluster to filter east-west Pod traffic. It cannot override or replace VPC firewall rules that operate at the Compute Engine VM (node) boundary.443 and 10250 using VPC firewall rules. Without an explicit VPC ingress allow rule for port 9443 from masterIpv4CidrBlock, packets from the API server are dropped at the node VM interface before reaching Pod NetworkPolicy enforcement.--master-authorized-networks) only govern IP-based control plane endpoints and do not restrict DNS-based endpoint traffic.NetworkPolicy) permitting TCP port 9443 from masterIpv4CidrBlock.--enable-private-nodes, --enable-private-endpoint, --enable-master-global-access, --enable-master-authorized-networks, and --master-authorized-networks=10.50.0.0/24. Add a VPC ingress firewall rule allowing TCP port 9443 from 10.50.0.0/24 to the cluster node target tags.This configuration enables private nodes, disables the external control plane endpoint, enables global access for the private endpoint, specifies 10.50.0.0/24 in --master-authorized-networks, and creates a VPC ingress firewall rule for TCP port 9443 sourced from 10.50.0.0/24.
--enable-authorized-networks-on-private-endpoint is omitted, GKE does not enforce the --master-authorized-networks list on the internal endpoint. Combined with --enable-master-global-access, any VM or workload in any VPC subnet across all Google Cloud regions can reach the control plane's internal IP address.kube-apiserver), not directly by the CI/CD runners in 10.50.0.0/24. Allowing ingress on TCP port 9443 from 10.50.0.0/24 leaves traffic from masterIpv4CidrBlock blocked by the default VPC ingress deny rule.--enable-authorized-networks-on-private-endpoint, the control plane accepts connections from all internal IP addresses in the cluster's VPC network./28 CIDR block (masterIpv4CidrBlock) as the source range.--enable-authorized-networks-on-private-endpoint were included in the cluster flags, and the VPC firewall rule specified masterIpv4CidrBlock as the source range instead of 10.50.0.0/24.A common misconception is that --enable-master-authorized-networks automatically locks down the internal endpoint once --enable-private-endpoint is set; in reality, you must explicitly enforce authorized networks on the private endpoint.
--enable-private-nodes, --enable-private-endpoint, --enable-master-global-access, --enable-master-authorized-networks, --master-authorized-networks=10.50.0.0/24, and --enable-authorized-networks-on-private-endpoint. Add a VPC ingress firewall rule (at lower precedence than GKE's default rules) allowing TCP port 9443 from the control plane's masterIpv4CidrBlock to the cluster node target tags.This configuration combines private GKE node and control plane isolation, global internal endpoint routing, internal authorized network enforcement, and a targeted VPC ingress firewall rule so that both cross-region CI/CD automation and Kubernetes admission webhooks function securely without exposing public IP endpoints.
--enable-private-nodes ensures worker nodes receive only internal RFC 1918 IP addresses from the subnet's primary range, eliminating direct internet exposure.--enable-private-endpoint disables the control plane's external IP-based endpoint. Pairing --enable-master-authorized-networks and --master-authorized-networks=10.50.0.0/24 with --enable-authorized-networks-on-private-endpoint enforces the CIDR allowlist directly on the internal endpoint, preventing unauthorized VPC subnets from accessing the Kubernetes API server.--enable-master-global-access allows internal IP addresses from other Google Cloud regions (us-east1) within the same VPC to reach the us-central1 control plane's private IP endpoint.9443 from masterIpv4CidrBlock to the cluster node target tags permits the control plane to reach the webhook Pods directly, while keeping GKE's automatically created rules intact by using lower precedence.us-central1.kubelet (10250) and control plane communications remain uninterrupted.This is the only configuration that simultaneously disables public IP access, enforces CIDR-based restrictions on the internal control plane endpoint across regions, and opens the exact backend Pod port (9443) from the control plane CIDR at the VPC layer.
--enable-private-nodes, --enable-private-endpoint, --enable-master-authorized-networks, --master-authorized-networks=10.50.0.0/24, and --no-enable-google-cloud-access. Add a VPC ingress firewall rule allowing TCP port 443 from the control plane's masterIpv4CidrBlock to the cluster node target tags.This configuration provisions a private GKE cluster with private nodes, disables the external IP-based control plane endpoint, enables authorized networks, and disables access from Google Cloud external IP addresses (--no-enable-google-cloud-access), while adding a VPC ingress firewall rule for TCP port 443.
--enable-master-global-access, the control plane's internal endpoint is only reachable from internal IP addresses in the same region as the cluster (us-central1). The CI/CD runners in us-east1 (10.50.0.0/24) cannot reach the internal endpoint.--enable-master-authorized-networks alone only restricts the external endpoint. Without --enable-authorized-networks-on-private-endpoint, any internal IP in us-central1 within the VPC can still access the control plane's internal endpoint.443 and 10250. However, VPC firewall rules evaluate the target Pod or node port (9443), not the Kubernetes Service port (443).--no-enable-google-cloud-access blocks Google Cloud external IP addresses (such as public IPs assigned to Compute Engine VMs or Cloud Run), which is redundant when --enable-private-endpoint is already active and does not restrict internal VPC traffic.--no-enable-google-cloud-access can take several hours to propagate.targetPort (9443).--enable-master-global-access and --enable-authorized-networks-on-private-endpoint were added to the cluster configuration, and the VPC firewall rule allowed TCP port 9443 instead of 443.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