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 uses a hub-and-spoke network topology where transit-vpc connects to your on-premises network (192.168.0.0/16) via Dedicated Interconnect and a Cloud Router in us-central1. You peered transit-vpc with gke-spoke-vpc, exporting custom routes from transit-vpc and importing them into gke-spoke-vpc.
A private GKE cluster in gke-spoke-vpc (us-central1) has a private control plane endpoint (172.16.0.0/28) and a node subnet (10.20.0.0/20). Requirements:
192.168.10.0/24) need highly available kubectl access to the control plane.gke-spoke-vpc are prohibited.What should you do?
transit-vpc to advertise 172.16.0.0/28 via custom route advertisement, and enable custom route export and import on both the transit-vpc-to-gke-spoke-vpc peering and the GKE control plane peering.This approach attempts to resolve hybrid reachability solely through BGP custom route advertisements on the transit-vpc Cloud Router combined with VPC Network Peering custom route exchange across two consecutive peering connections.
gke-spoke-vpc via VPC Network Peering. Because peering is strictly non-transitive, gke-spoke-vpc cannot re-advertise or propagate routes learned from the Google-managed control plane VPC (172.16.0.0/28) across a second peering link to transit-vpc.transit-vpc Cloud Router can advertise 172.16.0.0/28 to on-premises routers using custom advertisement mode, transit-vpc itself has no valid route to 172.16.0.0/28 in its VPC routing table and drops incoming packets.--export-custom-routes on the GKE peering connection only allows the control plane VPC to import static or dynamic routes from gke-spoke-vpc; it never exports the control plane's subnet route beyond gke-spoke-vpc.The Dedicated Interconnect VLAN attachments and Cloud Router were deployed directly inside gke-spoke-vpc rather than transit-vpc, placing the hybrid connection only one peering hop away from the GKE control plane.
transit-vpc, and add the transit-vpc proxy subnet CIDR to the GKE cluster's Master Authorized Networks.This design deploys an internal passthrough Network Load Balancer and a proxy MIG inside the central transit-vpc and updates the GKE cluster's Master Authorized Networks allowlist to permit the proxy subnet.
transit-vpc over Dedicated Interconnect, the proxy VMs reside in transit-vpc. To reach 172.16.0.0/28, proxy traffic must still cross two VPC peering hops (transit-vpc to gke-spoke-vpc, then gke-spoke-vpc to the Google-managed control plane VPC), which fails due to VPC Network Peering non-transitivity.transit-vpc IP ranges to Master Authorized Networks only configures an API server access control filter; it cannot inject routes for 172.16.0.0/28 into transit-vpc.transit-vpc suffers from the same routing limitation when targeting a private GKE control plane in a peered spoke VPC.The private GKE cluster were deployed directly inside transit-vpc instead of gke-spoke-vpc, or if transit-vpc and gke-spoke-vpc were connected via HA VPN tunnels rather than VPC Network Peering.
transit-vpc and gke-spoke-vpc with Network Connectivity Center (NCC) VPC spokes, attach the Dedicated Interconnect VLAN attachments as hybrid spokes, and connect directly to 172.16.0.0/28.This approach replaces the VPC Network Peering link between transit-vpc and gke-spoke-vpc with Network Connectivity Center (NCC) VPC spokes and registers the Dedicated Interconnect VLAN attachments as NCC hybrid spokes to enable dynamic route exchange across the hub.
172.16.0.0/28) resides in a Google-managed tenant VPC connected to gke-spoke-vpc via VPC Network Peering, 172.16.0.0/28 is not added to the NCC hub route table and remains unreachable directly from on-premises.transit-vpc and gke-spoke-vpc, an intermediary proxy or bastion host inside gke-spoke-vpc is still necessary to reach the peered GKE control plane endpoint.The on-premises CI/CD runners only needed direct access to workload VMs or Private Service Connect endpoints residing inside gke-spoke-vpc's own subnets rather than the peered GKE control plane.
gke-spoke-vpc, advertise the load balancer's subnet from the transit-vpc Cloud Router using custom route advertisement, and authorize the proxy subnet in Master Authorized Networks.This solution places a managed instance group (MIG) of TCP/HTTPS proxy instances (such as Envoy, HAProxy, or NGINX) behind an internal passthrough Network Load Balancer (Internal TCP/UDP Load Balancer) inside gke-spoke-vpc. The proxy terminates incoming kubectl API requests from on-premises and initiates new connections locally from within gke-spoke-vpc to the private GKE control plane endpoint.
transit-vpc, then crosses a single peering hop to the load balancer VIP in gke-spoke-vpc. The proxy VM then initiates a separate TCP session across a single peering hop from gke-spoke-vpc to 172.16.0.0/28.gke-spoke-vpc proxy subnet via Cloud Router custom route advertisement in transit-vpc enables on-premises routers to reach the load balancer VIP, while --export-custom-routes on transit-vpc provides the return path to 192.168.0.0/16.transit-vpc.gke-spoke-vpc for 192.168.10.0/24 and Google Cloud health check ranges.Unlike direct routing or placing proxies in transit-vpc, locating the internal TCP/UDP load balancer and proxy MIG inside gke-spoke-vpc isolates each side of the connection to a single peering hop, providing resilient, scalable hybrid access.
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