Prepare and test your skills

Prepare and test your skills

Free sample question with answers and detailed explanation below. The Associate Cloud Engineer practice bank has an additional 2,014 practice questions.Try 10 free realistic questions: instant score, no credit card, and no account required.
Your company runs a production application on a Google Kubernetes Engine (GKE) cluster in a node pool named pool-1. You create a new node pool named pool-2 with upgraded machine types and need to migrate your existing workloads to pool-2 before retiring pool-1.
You want to ensure that all workloads are safely rescheduled onto pool-2 with zero unexpected downtime before removing the old nodes.
Which sequence of actions should you take?
pool-1 to 0 and let GKE remove the nodes automatically.Adjusting the cluster autoscaler's maximum node boundary (--max-nodes) restricts the upper ceiling of nodes that the autoscaler can provision during high traffic or resource contention.
--max-nodes to 0 does not automatically scale down running instances if --min-nodes or current demand keeps nodes active; in fact, setting max nodes below current size is invalid or ineffective for draining.pool-2.You wanted to prevent a node pool from expanding further during an autoscaling event while allowing existing nodes to handle current traffic.
To decommission a node pool permanently, administrators must explicitly drain workloads and delete the node pool object rather than manipulating autoscaling limits.
pool-1, verify pods are running on pool-2, and then delete pool-1.In Kubernetes and Google Kubernetes Engine (GKE), cordoning marks a node as unschedulable to prevent new Pods from being placed on it, while draining gracefully evicts and terminates existing running Pods so that the Kubernetes control plane reschedules them onto other eligible nodes in the cluster.
kubectl cordon ensures no new incoming workloads land on pool-1. Running kubectl drain respects Pod Disruption Budgets (PDBs) and graceful termination periods, cleanly evicting Pods so they get scheduled onto available nodes in pool-2.pool-2 and the old nodes are confirmed empty, deleting pool-1 with gcloud container node-pools delete safely removes the underlying Compute Engine instances without data loss or interrupted requests.pool-2.
pool-1 directly using gcloud container node-pools update to match pool-2.In GKE, certain node pool properties (such as node auto-upgrade, auto-repair, and autoscaling boundaries) can be modified in-place on an existing node pool using the CLI or Google Cloud Console.
The operator were updating mutable pool properties such as enabling cluster autoscaling, updating node labels, or changing maintenance windows.
Because machine types are immutable, creating a new node pool and executing a cordon-and-drain migration is the standard method for resizing cluster nodes.
pool-1 immediately using gcloud container node-pools delete so that GKE automatically moves all pods to pool-2.Deleting a node pool directly via the Google Cloud CLI immediately signals GKE to tear down the managed instance groups and delete the virtual machine instances associated with that specific pool.
pool-2 lacks sufficient CPU, memory, or compatible taints/tolerations, evicting pods via direct deletion can leave critical workloads in an unschedulable Pending state.pool-2 encounters unexpected configuration issues.The cluster were running purely non-critical, fault-tolerant batch workloads or a disposable development environment where sudden termination and downtime carry no operational impact.
Always perform explicit node cordon and drain operations in production environments to maintain visibility and control over workload redistribution.
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