Kubernetes (advanced)plain Mode
Review all 15 questions for Kubernetes (advanced).Plain Q&A (No XP)
#1Multiple Choice
Which RBAC resource object binds a ClusterRole to a ServiceAccount scoped strictly within a single namespace?
#2Multiple Choice
What happens when a Pod exceeds its Memory Limit (`limits.memory`) in Kubernetes?
#3Multiple Choice
Which component in Kubernetes autoscaling architecture dynamically provisions physical worker nodes when Pods are in `Pending` state due to insufficient cluster resources?
#4Multiple Choice
What Kubernetes feature allows attaching custom metadata key-value taints to Nodes so only Pods with matching tolerations can be scheduled on them?
#5Multiple Choice
Why should Horizontal Pod Autoscaler (HPA) and Vertical Pod Autoscaler (VPA) not be configured to autoscale on the same metric (e.g. CPU) simultaneously?
#6Multiple Choice
What API group and resource provides admission control webhook validation to intercept and mutate Kubernetes API requests before etcd persistence?
#7Scenario
A newly deployed Pod remains stuck in `Pending` state. Running `kubectl get pods` shows 0/1 READY. `kubectl describe pod` events log shows: '0/4 nodes are available: 4 Insufficient memory.' The node memory total is 64 GB, but actual memory usage is only 20 GB. Why is scheduling failing?
#8Scenario
You are setting up dedicated GPU worker nodes in your cluster for ML training. You want to ensure non-ML general workloads never run on these expensive GPU nodes. What two Kubernetes configuration steps are required?
#9Scenario
An attacker exploits an RCE in your web application container. They attempt to access the Cloud Provider Metadata API (`http://169.254.169.254`) to steal IAM node instance credentials. What Kubernetes resource prevents this Pod from reaching that internal IP?
#10Scenario
A node drops offline due to a network partition. How long does the Kubernetes `node-lifecycle-controller` wait by default before marking the node `NotReady` and evicting its pods to other nodes?
#11Code Fill
Complete the Pod securityContext setting to enforce that container processes can never escalate privileges via setuid:
spec:
containers:
- name: app
securityContext:
allowPrivilegeEscalation: ___#12Code Fill
Complete the Pod spec key used to define budget guarantees preventing accidental node drains from taking down all replicas:
apiVersion: policy/v1
kind: ___
metadata:
name: api-pdb
spec:
minAvailable: 2#13Code Fill
Complete the command to cordon a node so no new pods can be scheduled on it prior to maintenance:
kubectl ___ node-01#14Open Ended
Compare Kubernetes eBPF-based CNI network plugins (like Cilium) with traditional iptables/kube-proxy network plugins.
#15Open Ended