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 manage an HA VPN connection between a Google Cloud VPC and an on-premises VPN gateway with two tunnels (tunnel-if0 and tunnel-if1) and dynamic BGP routing. Users report brief application stalls, and your diagnostics show:
gcloud compute vpn-gateways get-status reports CONNECTION_REDUNDANCY_MET.resource.type="vpn_gateway") shows Received SA_DELETE on tunnel-if0 right before it drops for several seconds and re-establishes, with no Cloud Router maintenance logs.tunnel_established hits 0, BGP shows The VPN Tunnel is not in Established state., and network/received_packets_count logs sa_expired drops.tunnel-if1 triggers egress exceeds_mtu drops in network/sent_packets_count.Which root cause analysis and remediation plan resolves these issues?
tunnel-if0 using IKEv1 with an identical cipher list to prevent Received SA_DELETE renegotiation drops, and run ping tests from an on-premises host to the Cloud VPN gateway external IP address to isolate the exceeds_mtu drops.IKEv1 is an older IPsec key-exchange protocol requiring an exact match of cipher suites between both gateways, whereas IKEv2 auto-negotiates when at least one common supported cipher is proposed. Pinging a VPN gateway external IP checks public Layer 3 reachability to the gateway's WAN interface.
no-proposal-chosen errors.exceeds_mtu unresolved: Neither changing the IKE version nor pinging an external IP prevents Google Cloud VMs from sending oversized egress packets with the do not fragment bit set.sa_expired drops caused by peer SA deletion.ping or iperf diagnostics require internal VM-to-host testing with appropriate ingress/egress ICMP firewall rules on both sides.An IKEv1-only legacy peer failed initial Phase 1 negotiation due to cipher mismatches and you needed to verify basic WAN reachability to the gateway's public IP before troubleshooting tunnel establishment.
Maintenance of router task events, and enable Bidirectional Forwarding Detection (BFD) on tunnel-if0 and tunnel-if1 to eliminate sa_expired and exceeds_mtu drops.BGP graceful restart preserves learned routing tables during control-plane software restarts—such as Maintenance of router task events—so data-plane forwarding continues uninterrupted while BGP sessions re-establish. Bidirectional Forwarding Detection (BFD) provides sub-second liveness detection at the routing layer to accelerate BGP failover when a physical or logical link fails.
Maintenance of router task entries, meaning the BGP session transitions to The VPN Tunnel is not in Established state. are secondary symptoms of the underlying IPsec tunnel going down rather than a Cloud Router restart.Received SA_DELETE and sa_expired packet drops indicate that the on-premises peer is deleting its Phase 2 Security Association (SA) instead of rekeying proactively. BFD only detects link failure faster; it cannot prevent the peer from tearing down the IPsec tunnel.exceeds_mtu drops recorded on network/sent_packets_count, which occur when Google Cloud VMs transmit oversized UDP or ESP packets with the do not fragment bit set.DOWN (The VPN Tunnel is not in Established state.).You observed Maintenance of router task: BGP sessions will restart entries in Cloud Router logs with premature route withdrawal during planned maintenance, rather than Received SA_DELETE and exceeds_mtu metrics.
do not fragment settings so egress UDP/ESP traffic stays within the tunnel MTU.Pre-expiration Phase 2 SA rekeying negotiates a replacement Security Association (SA) before the active SA expires, maintaining uninterrupted encrypted forwarding. Egress MTU compliance ensures packets sent by Google Cloud VMs fit within the Cloud VPN tunnel Maximum Transmission Unit (MTU) after IPsec encapsulation overhead.
Received SA_DELETE and sa_expired drops: Cloud VPN rekeys proactively by default. Seeing Received SA_DELETE immediately before a brief disconnection—alongside sa_expired ingress drops and tunnel_established falling to 0—confirms that the on-premises gateway is deleting the existing SA before negotiating a new one. Enabling proactive rekeying on the peer with a matching Phase 2 lifetime and a supported Diffie-Hellman (DH) group prevents the tunnel and its BGP session (The VPN Tunnel is not in Established state.) from flapping.exceeds_mtu egress drops: During failover to tunnel-if1, network/sent_packets_count reports exceeds_mtu when the first fragment of a UDP or ESP egress packet from a Google Cloud VM exceeds the tunnel MTU while the do not fragment (DF) bit is set. Lowering the VM or application MTU or allowing fragmentation prevents the Cloud VPN gateway from dropping those packets.resource.type="vpn_gateway" pinpoints whether tunnel drops stem from peer SA_DELETE behavior rather than Cloud Router maintenance.network/received_packets_count (sa_expired) and network/sent_packets_count (exceeds_mtu) separates peer SA mismatches from VM egress framing issues.CONNECTION_REDUNDANCY_MET with active/active tunnels ensures failover validation succeeds without packet loss.Unlike tuning Cloud Router timers, DPD settings, or IKEv1 ciphers, fixing peer Phase 2 rekeying and VM egress MTU directly targets both proven root causes shown in the logs and metrics.
tunnel-if0 fails over faster before SA deletion, and modify the local and remote traffic selectors on tunnel-if1 to resolve the exceeds_mtu egress drops.Dead Peer Detection (DPD) is an RFC 3706 IKE mechanism that probes an unresponsive VPN peer and removes routes pointing to an unhealthy tunnel so traffic fails over to remaining healthy tunnels. Traffic selectors define the permitted local and remote IP CIDR ranges allowed to traverse a VPN tunnel.
SA_DELETE instead of rekeying.exceeds_mtu drops.0.0.0.0/0 for both local and remote traffic selectors, relying on Cloud Router BGP routes rather than selector-based filtering.sa_expired and exceeds_mtu reflect SA lifecycle mismatches and oversized do not fragment packets, not IP prefix filtering.Traffic failed completely between specific VM and on-premises subnets on a Classic VPN tunnel due to narrow encryption domains, rather than experiencing Received SA_DELETE rekeying gaps and exceeds_mtu packet drops on HA VPN.
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