cluster-status
Check Kubernetes cluster health and status. Use when checking nodes, pods, deployments, or overall cluster health.
When & Why to Use This Skill
This Claude skill provides a comprehensive toolkit for monitoring Kubernetes cluster health and operational status. It automates the retrieval of critical metrics for nodes, pods, and deployments across specific namespaces like vLLM and Temporal. By integrating kubectl diagnostics with Flux GitOps tracking, it enables SREs and DevOps engineers to maintain high availability, visualize resource utilization, and ensure all infrastructure components are in a 'Ready' state.
Use Cases
- Real-time Cluster Health Auditing: Quickly verify the status of nodes and pods across multiple namespaces to ensure system-wide stability.
- DevOps Troubleshooting: Identify and diagnose common container issues such as ImagePullBackOff, CrashLoopBackOff, or resource-related Pending states.
- GitOps Reconciliation Tracking: Monitor Flux CD status to ensure that the live cluster state is synchronized with the desired configuration in Git.
- Resource Capacity Planning: Use node resource usage data and event logs to detect bottlenecks and optimize hardware allocation within the cluster.
| name | cluster-status |
|---|---|
| description | Check Kubernetes cluster health and status. Use when checking nodes, pods, deployments, or overall cluster health. |
Cluster Status Check
Check the health and status of the Kubernetes cluster.
Instructions
Quick Status Overview
KUBECONFIG=/home/al/.kube/config
echo "=== Nodes ==="
kubectl get nodes -o wide
echo ""
echo "=== AI Agents Namespace ==="
kubectl get pods -n ai-agents
echo ""
echo "=== vLLM Namespace ==="
kubectl get pods -n vllm
echo ""
echo "=== Temporal Namespace ==="
kubectl get pods -n temporal
Detailed Node Health
KUBECONFIG=/home/al/.kube/config
# Node resource usage
kubectl top nodes
# Node conditions
kubectl get nodes -o jsonpath='{range .items[*]}{.metadata.name}{"\t"}{.status.conditions[-1].type}={.status.conditions[-1].status}{"\n"}{end}'
Flux GitOps Status
KUBECONFIG=/home/al/.kube/config flux get all -A
Recent Events
KUBECONFIG=/home/al/.kube/config kubectl get events --sort-by='.lastTimestamp' -A | tail -20
Problem Pods
KUBECONFIG=/home/al/.kube/config kubectl get pods -A | grep -v Running | grep -v Completed
Health Indicators
- All nodes should be
Ready - Critical pods should be
Running - No pending or failed pods
- Flux reconciliation should be successful
Common Issues
- ImagePullBackOff: Image not pushed to registry
- CrashLoopBackOff: Application crash, check logs
- Pending: Insufficient resources or scheduling issues
- NotReady nodes: Node health issue, check kubelet