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 configuring control plane access for a GKE Standard cluster in a hybrid development environment with the following requirements:
Which configuration should you implement?
VPC firewall rules filter ingress and egress traffic entering or leaving Compute Engine VM network interfaces inside a customer's VPC network. While GKE automatically creates essential VPC firewall rules so worker nodes and Pods can communicate with the control plane, those rules apply strictly at the worker node VM boundary within the customer VPC.
--enable-master-authorized-networks and --no-enable-google-cloud-access, the public IPv4 endpoint remains reachable from any IP address on the internet and from any Google Cloud customer's external IP.container.googleapis.com and kubernetesmetadata.googleapis.com in the service perimeter's restricted services list.gcloud container clusters get-credentials automatically configures kubectl to use the IP-based endpoint unless --dns-endpoint is explicitly passed. Without --dns-endpoint, developers bypass the DNS-based endpoint and its VPC Service Controls enforcement.0.0.0.0/0 deny rules in a VPC can inadvertently interfere with GKE node health checks, kubelet traffic, or internal Pod communication if priorities override GKE's system rules.--no-enable-google-cloud-access prevents arbitrary external IPs assigned to other Google Cloud tenants from accessing the IP-based endpoint.Instead of configuring a VPC firewall rule, you enabled --enable-master-authorized-networks with --no-enable-google-cloud-access, added kubernetesmetadata.googleapis.com to the VPC Service Controls perimeter, and instructed developers to run get-credentials --dns-endpoint.
Passing --enable-private-endpoint disables the GKE control plane's external IP-based endpoint while keeping only the internal IP-based endpoint active within the cluster's VPC network. This configuration is commonly used in strict private cluster deployments to eliminate all public IPv4 access to the Kubernetes API server.
--enable-private-endpoint removes external access to the control plane's public IPv4 address, the external CI/CD pipeline can no longer connect to the public IPv4 endpoint, regardless of whether its static public CIDR block is listed in --master-authorized-networks.--enable-private-endpoint, clients outside the VPC (or outside peered/interconnected hybrid networks) cannot reach the IP-based endpoint at all without deploying a bastion host or switching to the DNS-based endpoint.--no-enable-ip-access) and using only the DNS-based endpoint.The CI/CD pipeline had no requirement to connect via a public IPv4 endpoint and instead used the DNS-based endpoint (authenticated via IAM or enabled Kubernetes credentials) or ran inside the VPC/on-premises network over the internal endpoint.
Google Kubernetes Engine (GKE) allows a cluster control plane to expose both a DNS-based endpoint (*.gke.goog) and IP-based endpoints (internal and external IPv4 addresses) simultaneously. By pairing IAM-authenticated DNS access with strictly scoped authorized networks on the IP-based endpoints, network engineers can support legacy or static-IP integrations alongside modern zero-trust developer workflows.
--enable-ip-access without --enable-private-endpoint keeps the external IPv4 endpoint active, while --enable-master-authorized-networks and --master-authorized-networks restrict external IPv4 access strictly to the CI/CD platform's static CIDR block.--enable-dns-access provides an immutable FQDN served through the Google Front End (GFE) that authenticates users via IAM (container.clusters.connect) without requiring bastion hosts or constant CIDR updates. Adding both container.googleapis.com and kubernetesmetadata.googleapis.com to a VPC Service Controls perimeter enforces context-aware perimeter policies across all GKE API operations. Passing --dns-endpoint to gcloud container clusters get-credentials overrides the CLI's default preference for the IP-based endpoint.--enable-master-global-access allows internal IP addresses in any Google Cloud region to reach the control plane's internal endpoint, and Cloud Router custom route advertisements propagate the control plane's internal CIDR to on-premises hosts over Cloud Interconnect.--no-enable-google-cloud-access blocks external IP addresses belonging to other Google Cloud customers (such as arbitrary Compute Engine VMs or Cloud Run instances) from reaching the control plane.--no-enable-google-cloud-access are not instantaneous and can take several hours for GKE to fully propagate and enforce.--enable-authorized-networks-on-private-endpoint is also enabled, VPC and hybrid internal ranges can reach the internal endpoint without being explicitly listed in --master-authorized-networks.This architecture cleanly separates human access from machine-to-machine IPv4 access. Unlike disabling the external endpoint (which breaks the CI/CD requirement) or relying on VPC firewall rules (which do not protect the Google-managed control plane), it enforces least-privilege controls at every boundary.
When IP-based endpoints and authorized networks are enabled, allowing Google Cloud external IP addresses permits connections from external IPs owned by Google Cloud across any customer's Compute Engine VMs, Cloud Run services, or Cloud Run functions. Meanwhile, running gcloud container clusters get-credentials without --dns-endpoint configures kubectl to target the cluster's IP-based endpoint whenever IP access is active.
get-credentials defaults to the IP-based endpoint, developers connecting from dynamic home IP addresses attempt to reach the public IPv4 endpoint instead of the DNS-based endpoint. Their dynamic IPs are rejected by authorized networks unless manually allowlisted on every IP change.get-credentials settings never route through the DNS-based endpoint and therefore do not benefit from VPC Service Controls perimeter enforcement.--dns-endpoint to gcloud container clusters get-credentials is mandatory whenever IP-based access is enabled alongside DNS access so clients resolve the immutable *.gke.goog FQDN.You passed --no-enable-google-cloud-access to block external IPs belonging to other Google Cloud customers and required developers to run gcloud container clusters get-credentials --dns-endpoint.
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