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 runs latency-sensitive Containerized Network Functions (CNFs) on a GKE Standard cluster. The existing C2 node pool uses Container-Optimized OS with VirtIO interfaces. You must upgrade the networking architecture to meet these requirements:
Which configuration should you implement?
Running gcloud container node-pools update with --enable-gvnic modifies the underlying node template of an existing GKE Standard node pool to switch nodes from VirtIO to Google Virtual NIC (gVNIC), forcing GKE to replace every VM instance in the pool.
--enable-gvnic on an existing node pool is a manual change that immediately begins recreating nodes using the configured node upgrade strategy (such as surge or blue-green upgrades). GKE performs this recreation regardless of active maintenance policies.--enable-gvnic updates.netdevice mode to bypass GKE Dataplane V2.Immediate workload disruption from node recreation were acceptable outside maintenance windows, no additional physical vNICs were needed on the nodes, and Pods did not require multi-network netdevice passthrough.
Provisioning a parallel GKE Standard node pool with Google Virtual NIC (gVNIC), per-VM Tier_1 networking, and GKE multi-network support for Pods in netdevice mode establishes a high-throughput, low-latency host and Pod data path. Replacing the VirtIO pool via a side-by-side pool migration acts as a controlled blue-green transition.
--enable-gvnic on Container-Optimized OS nodes is a mandatory prerequisite for Tier_1 networking on C2 (up to 100 Gbps) and C3 (up to 200 Gbps) instances. Attaching the secondary interface to a separate VPC configured with MTU 8896 maximizes packet efficiency and reduces per-packet CPU overhead.netdevice mode passes the secondary VM NIC directly into the CNF Pod. This completely bypasses Kubernetes networking and GKE Dataplane V2 for the data plane while keeping standard control-plane and cluster communication on the primary interface.--enable-gvnic on an active pool, letting you cordon and drain workloads cleanly according to your operational schedule.Unlike in-place node pool updates that ignore maintenance policies or container-namespace workarounds that still traverse host networking, provisioning a dedicated netdevice multi-network node pool delivers deterministic ultra-low latency and full control over workload cutover.
In GKE Autopilot clusters running version 1.30.2-gke.1023000 or later, nodes automatically include Google Virtual NIC (gVNIC). Moving a secondary host NIC directly into an isolated Linux network namespace (ip netns) is a manual host-level pattern used in some VM or unmanaged environments to give an application exclusive access to a NIC.
CAP_SYS_ADMIN Linux capability and a privileged container. This configuration is incompatible with GKE Autopilot, which restricts privileged containers and low-level host network manipulation for security and managed operations.ip link namespace commands.netdevice orchestration: Rather than relying on manual ip netns scripts inside privileged containers, GKE provides multi-network support for Pods in netdevice mode on Standard clusters.CAP_SYS_ADMIN and privileged host-network access.ip link set eth1 netns is used on Standard VMs, other applications outside that namespace lose access to the interface.The workloads ran on GKE Standard rather than Autopilot, privileged containers with CAP_SYS_ADMIN were permitted by security policy, and manual network namespace isolation were preferred over GKE multi-network netdevice mode.
This approach attempts to combine two gVNIC interfaces on the same VPC using Linux kernel bonding for link aggregation while using the networking.gke.io/default-interface Pod annotation to redirect default container traffic to eth1.
networking.gke.io/default-interface: 'eth1' only modifies the default route inside the Pod's Kubernetes container network namespace. Traffic still traverses the container network stack rather than passing the host NIC directly to the Pod via netdevice mode to bypass Kubernetes networking and GKE Dataplane V2.eth0) to remain dedicated to internal Kubernetes cluster communication.Each gVNIC interface were connected to a separate VPC without kernel bonding, and the CNF workloads only needed default-route redirection inside the container namespace rather than direct netdevice passthrough.
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