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 deployed a multi-NIC NGFW appliance in a managed instance group (MIG) behind internal passthrough Network Load Balancers (ILBs) to inspect inter-VPC and internet-bound traffic:
nic0 (untrust) in untrust-vpc handles internet egress.nic2 in vpc-a (10.10.0.0/16) and nic3 in vpc-b (10.20.0.0/16) sit behind regional ILBs (ilb-a and ilb-b).During validation, ILB health checks fail and all spoke traffic is dropped. You discover canIpForward is disabled on the appliance template, firewall rules only allow intra-subnet traffic, and vpc-a and vpc-b have untagged 0.0.0.0/0 and inter-VPC static routes pointing to the ILBs.
Which configuration changes resolve these issues?
Configuring IP forwarding (canIpForward), network-tagged static routes, and VPC firewall rules across all attached interfaces enables a multi-NIC network virtual appliance (NVA) to operate as an inline Layer-3/Layer-7 security bridge between multiple VPC networks and the internet.
canIpForward): By default, Compute Engine blocks packets whose source or destination IP does not match an interface on the VM (anti-spoofing). Because canIpForward can only be set when a VM is created, updating the instance template and performing a rolling replace disables these anti-spoofing checks so the NGFW can receive and forward packets belonging to vpc-a and vpc-b workloads.inspect-workload) to workload VMs and referencing that tag on the custom inter-VPC and 0.0.0.0/0 static routes steers workload traffic to the internal passthrough Network Load Balancers (ILBs) while preventing the NGFW's own nic2 and nic3 interfaces from matching those routes and looping traffic back to the ILBs.10.10.0.0/16 and 10.20.0.0/16, internet egress (0.0.0.0/0) on untrust-vpc, and ingress from the load balancer health-check probe ranges (35.191.0.0/16 and 130.211.0.0/22).nic2, nic3, or nic0.untrust-vpc, vpc-a, and vpc-b traffic planes at each vNIC.canIpForward retains original workload IP addresses for deep packet inspection and security logging.Unlike modifying running VMs in place (which fails for canIpForward), referencing secure tags in static routes (which is unsupported), or relying on out-of-band Packet Mirroring, this solution resolves every forwarding, routing-loop, and firewall-plane blocker in a single coherent design.
This option attempts to enable IP forwarding on existing VMs without recreating them, use Resource Manager secure tags as route filters on static routes, and apply hierarchical firewall policies to permit transit and health-check traffic.
canIpForward in place: In Google Cloud, IP forwarding is an immutable VM-creation property. You cannot enable canIpForward on running Compute Engine instances; you must update the instance template and recreate the VMs. Without this change, the VPC data plane drops all transit packets due to source and destination anti-spoofing checks.canIpForward or modifying attached network interfaces always requires replacing the VM instance.The solution recreated the NGFW VMs from an updated instance template with canIpForward enabled, used network tags on the static routes in vpc-a and vpc-b, and reserved secure tags solely for the firewall policy rules.
A common misconception is treating secure tags as a drop-in replacement for network tags across all networking features. Always distinguish between routing (requires network tags) and firewall filtering (supports either, with secure tags preferred for IAM governance).
This option attempts to add a new network interface to the running NGFW VMs, toggle IP forwarding at the instance group level, and use Packet Mirroring alongside health-check firewall rules to inspect inter-VPC and internet-bound traffic.
canIpForward must be configured in the instance template at VM creation time, not as an instance group setting.35.191.0.0/16 and 130.211.0.0/22) leaves 10.10.0.0/16, 10.20.0.0/16, and internet egress traffic blocked by VPC firewall rules, while untagged routes remain misconfigured.canIpForward to sit directly in the data path.The requirement were strictly passive, out-of-band traffic monitoring using a collector instance group rather than inline NGFW routing and policy enforcement.
This approach properly recreates the NGFW VMs with IP forwarding (canIpForward) enabled and configures basic VPC firewall rules for inter-VPC (10.10.0.0/16, 10.20.0.0/16) and internet egress (0.0.0.0/0) traffic, but leaves the static routes untagged and omits load balancer health-check firewall rules.
35.191.0.0/16 and 130.211.0.0/22). Without explicit ingress allow firewall rules permitting these ranges on nic2 and nic3, probes are dropped by the implied deny-ingress rule and the ILBs mark all NGFW backends unhealthy.0.0.0.0/0 and inter-VPC static routes untagged applies them to every VM in vpc-a and vpc-b, including the NGFW's own nic2 and nic3 interfaces. When the appliance forwards traffic or returns health-check responses out those interfaces, untagged routes can steer traffic back toward the ILB, creating a routing loop.priority 65535) that blocks health-check probes unless explicitly permitted.35.191.0.0/16 and 130.211.0.0/22 regardless of the VPC's internal CIDR.A network tag were applied to the workload VMs and custom static routes in vpc-a and vpc-b to exclude the NGFW interfaces, and ingress firewall rules explicitly allowed 35.191.0.0/16 and 130.211.0.0/22 on the health-check port.
Even when an NVA has canIpForward enabled and transit rules configured, omitting health-check probe ranges keeps the ILB forwarding rule inactive and prevents traffic from reaching the appliance.
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