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 GKE cluster in us-central1 using IP-based control plane endpoints. Your security team requires strict zero-trust network boundaries for administrative access:
10.20.0.0/24) in us-east1 and an on-premises network (172.16.10.0/24) connected via Cloud Interconnect with custom route advertisements already configured.Which set of gcloud flags should you apply when configuring the GKE cluster?
Combining private endpoint isolation, Master Authorized Networks, internal endpoint enforcement, control plane global access, and Google Cloud external IP denial establishes a strict zero-trust boundary around an IP-based GKE control plane.
--enable-private-endpoint): Sets enablePublicEndpoint to false, preventing all client connections to the control plane over the public internet.--enable-master-authorized-networks and --master-authorized-networks=10.20.0.0/24,172.16.10.0/24): Enables the IP-based firewall and explicitly permits only the us-east1 bastion subnet and the on-premises operations network.--enable-authorized-networks-on-private-endpoint): Sets privateEndpointEnforcementEnabled to true, blocking unauthorized VPC subnets outside the cluster's Pod, Service, and node ranges from reaching the internal endpoint.--enable-master-global-access): Allows internal IP traffic from the us-east1 bastion subnet—and hybrid traffic routed across regions—to reach the internal endpoint in us-central1.--no-enable-google-cloud-access): Sets gcpPublicCidrsAccessEnabled to false, ensuring external Google Cloud IP addresses cannot access the control plane.Every flag in this configuration addresses a distinct layer of GKE control plane isolation. Omitting internal endpoint enforcement leaves the VPC unrestricted, omitting global access blocks cross-region subnets, and omitting --enable-private-endpoint leaves the public endpoint exposed.
This configuration disables the external control plane endpoint, activates Master Authorized Networks with the specified CIDR blocks, enables control plane global access, and denies Google Cloud external IP addresses—but omits --enable-authorized-networks-on-private-endpoint.
In GKE, --enable-master-authorized-networks alone does not enforce the CIDR allowlist against the control plane's internal IP endpoint unless private endpoint enforcement (privateEndpointEnforcementEnabled) is explicitly enabled. Without --enable-authorized-networks-on-private-endpoint, the internal endpoint remains accessible from any internal IP address in the cluster's VPC network. Combined with --enable-master-global-access, workloads in unrelated VPC subnets across any Google Cloud region can reach the Kubernetes API server, failing the zero-trust requirement to block unauthorized VPC subnets.
--enable-master-authorized-networks creates a false sense of microsegmentation inside shared or multi-subnet VPCs.This configuration would satisfy the requirements only if all subnets within the cluster's VPC network were trusted to access the control plane, or if --enable-authorized-networks-on-private-endpoint were added to enforce the CIDR allowlist on the internal interface.
This configuration enables Master Authorized Networks, restricts the internal endpoint to the listed CIDRs, enables control plane global access, and denies Google Cloud external IP addresses—but omits --enable-private-endpoint.
Omitting --enable-private-endpoint leaves the control plane's external (public) IP endpoint active (enablePublicEndpoint: true). Although the allowlist contains only RFC 1918 private CIDR ranges (10.20.0.0/24 and 172.16.10.0/24), the external endpoint remains provisioned and accessible to preset Google-reserved IP addresses, directly violating the requirement to completely disable the external IP endpoint. In strict regulatory and zero-trust architectures, filtering a public endpoint is not equivalent to disabling it.
--master-authorized-networks, remote internet access to the external endpoint is immediately possible.This configuration would be appropriate only if the organization intended to retain the external control plane endpoint for remote administration from allowlisted public IP addresses rather than disabling public endpoint access altogether.
This configuration disables the public control plane endpoint, enforces Master Authorized Networks on both external and internal interfaces, and blocks Google Cloud external IP addresses—but omits --enable-master-global-access.
By default, a GKE control plane's internal IP endpoint is regionally scoped, meaning it is reachable only from internal IP addresses located in the same Google Cloud region as the cluster (us-central1). Even though 10.20.0.0/24 is included in --master-authorized-networks, the bastion host resides in us-east1. Without --enable-master-global-access (globalAccess: true), packets from us-east1 cannot reach the internal endpoint in us-central1, breaking administrative kubectl access from the bastion subnet.
us-central1, on-premises traffic from 172.16.10.0/24 is similarly unable to reach the private endpoint without global access.us-central1 resources.This configuration would be sufficient only if the administrative bastion subnet and on-premises interconnect attachment resided exclusively in us-central1, or if --enable-master-global-access were included to allow cross-region private reachability.
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