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 Router Appliance (RA) connectivity in Google Cloud Network Connectivity Center (NCC) to link an on-premises network with multiple VPC spokes. You deploy two RA VMs (ra-vm-1 in us-central1-a and ra-vm-2 in us-central1-b) in the same subnet (10.10.0.0/24) of transit-vpc and attach them as a Router appliance spoke to an NCC hub alongside vpc-spoke-a and vpc-spoke-b.
You need to satisfy the following requirements:
us-central1 and both RA VMs with ECMP load balancing.172.16.50.0/24) through the RA VMs, and prevent 10.30.10.0/24 from being advertised from vpc-spoke-b to the RA VMs.Which configuration and verification approach should you implement?
10.10.0.0/24, establish four BGP sessions (two per RA VM) allowing TCP port 179 with identical MEDs, configure custom learned routes on the BGP sessions paired with static routes on the RA VMs for 172.16.50.0/24, apply a BGP route policy (or VPC spoke exclude-export filter) to drop 10.30.10.0/24, and verify the NCC hub route table and routers.getRouterStatus.This architecture combines redundant Cloud Router BGP peering, custom learned routes, BGP route policies (or VPC spoke export filters), and NCC control-plane verification to connect on-premises networks to multiple VPC spokes through third-party virtual appliances.
10.10.0.0/24 and permitting TCP port 179 allows each RA VM to establish two BGP sessions—one to each Cloud Router interface, creating four sessions total. Advertising identical prefixes with matching Multi-Exit Discriminator (MED) values across both RA VMs enables Google Cloud to apply equal-cost multipath (ECMP) routing.172.16.50.0/24 without BGP advertisement: Defining 172.16.50.0/24 as a custom learned route on the Cloud Router BGP sessions causes Cloud Router to program dynamic routes into the VPC as if learned from the RA VMs. Pairing this with a static route on each RA VM pointing to the external next hop ensures the appliances know how to forward data-plane packets for 172.16.50.0/24.10.30.10.0/24 from vpc-spoke-b: Applying an exclude-export range on vpc-spoke-b prevents 10.30.10.0/24 from entering the NCC hub route table, while a BGP route policy on Cloud Router can selectively drop or modify specific prefixes before advertisement to the RA peers.routers.getRouterStatus verifies live BGP session establishment, learned routes, and advertised routes.ra-vm-1 and ra-vm-2 in separate zones with two BGP sessions each protects against both zonal VM failures and Cloud Router task restarts.Unlike link-local peering, single-session topologies, or unequal-cost routing designs, this configuration adheres strictly to Router Appliance subnet peering rules, guarantees ECMP load balancing, and provides deterministic prefix filtering and verification across the entire hybrid topology.
10.10.0.0/24, establish two BGP sessions (one per RA VM, alternating interfaces) with unequal MED values, configure custom learned routes on Cloud Router for 172.16.50.0/24 without adding static routes on the RA VMs, switch Cloud Router to CUSTOM advertisement mode with ALL_SUBNETS to omit 10.30.10.0/24, and verify learned routes in the Cloud Router status.This approach sets up partial BGP peering with one BGP session per RA VM, assigns unequal MED values, configures custom learned routes solely on Cloud Router, and relies on custom advertisement mode with ALL_SUBNETS enabled to control prefix propagation.
ra-vm-1 and ra-vm-2, forcing active/passive forwarding toward whichever session has the lower MED.172.16.50.0/24: Fails because configuring a custom learned route on Cloud Router only instructs the VPC to route 172.16.50.0/24 traffic to the RA VMs. Without a matching static route configured inside each RA VM pointing toward the external network, the appliances will drop or misroute the forwarded traffic.10.30.10.0/24 from vpc-spoke-b: Fails because enabling ALL_SUBNETS (advertisedGroups: [ALL_SUBNETS]) under CUSTOM advertisement mode causes Cloud Router to continue advertising all subnets visible to it according to the VPC's dynamic routing mode, including 10.30.10.0/24.172.16.50.0/24 were added on both RA VMs, and either a BGP route policy or a VPC spoke exclude-export filter was used instead of ALL_SUBNETS to drop 10.30.10.0/24.169.254.0.0/16 link-local addresses, establish four IPv4 BGP sessions with MP-BGP enabled for dual-stack route exchange, advertise 172.16.50.0/24 as a custom learned route on the RA VMs only, apply an exclude-export filter on the Router appliance spoke to block 10.30.10.0/24, and verify using routers.getRouterStatus.This configuration attempts to apply BGP addressing and protocol patterns commonly used for Cloud VPN and Cloud Interconnect VLAN attachments—specifically 169.254.0.0/16 link-local peering addresses and Multiprotocol BGP (MP-BGP)—to a Network Connectivity Center (NCC) Router Appliance deployment, while misplacing where custom learned routes and VPC spoke export filters are applied.
169.254.x.x link-local addresses. Both the Cloud Router interfaces and the RA VM interfaces must use RFC 1918 internal IP addresses within the same VPC subnet (10.10.0.0/24). Additionally, IPv6 BGP peering and IPv6 route exchange (including MP-BGP) are unsupported on Router Appliance.172.16.50.0/24 via custom learned routes: Fails because custom learned routes must be configured on the Cloud Router BGP session so Cloud Router injects dynamic routes into the VPC as if learned from the RA peer, alongside static routes on the RA VMs pointing toward the external network.10.30.10.0/24 from vpc-spoke-b: Fails because exclude-export IP address range filters are configured on VPC spokes (vpc-spoke-b) to control which VPC subnets enter the NCC hub route table, not on Router appliance spokes.include-export and exclude-export filters govern subnet advertisement from VPC spokes into the NCC hub route table.169.254.0.0/16 addresses and support MP-BGP), and the export filters and custom learned routes were configured on the appropriate VPC spoke and Cloud Router BGP sessions.A common misconception is treating Router Appliance BGP interfaces like VLAN attachment or VPN tunnel interfaces. Because RA VMs reside directly inside a VPC subnet, Cloud Router communicates with them over standard RFC 1918 subnet routing rather than point-to-point link-local tunnels.
10.10.0.0/24, establish four BGP sessions (two per RA VM) with different AS path lengths and matching MEDs, configure 172.16.50.0/24 using Cloud Router custom advertised ranges, attach an HA VPN spoke advertising the same on-premises prefixes for additional ECMP capacity, filter 10.30.10.0/24 with an ingress firewall rule on the RA VMs, and verify the NCC hub route table.This design attempts to achieve high-capacity hybrid routing by mixing Router Appliance spokes and HA VPN spokes for the same on-premises prefixes, modifying AS path lengths alongside matching MEDs, using custom advertised ranges for an inbound on-premises prefix, and relying on a VM firewall rule to suppress route advertisement.
172.16.50.0/24 from on-premises: Fails because Cloud Router custom advertised ranges advertise VPC-side prefixes outbound to BGP peers; they do not teach the VPC how to reach remote on-premises prefixes via custom learned routes.10.30.10.0/24 from vpc-spoke-b: Fails because an ingress firewall rule on the RA VMs only filters data-plane packets arriving at the VM NICs—it does not prevent Cloud Router from advertising 10.30.10.0/24 in BGP updates.exclude-export filters, not Compute Engine firewall rules.172.16.50.0/24, and a BGP route policy or exclude-export filter was applied to drop 10.30.10.0/24.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