Kubernetes Health Check Skill

Kubernetes Health Check

Purpose

Perform comprehensive health checks on Kubernetes clusters to ensure readiness for application deployments. Validates cluster connectivity, resource availability, and essential components.

Prerequisites

kubectl installed and configured
Access to Kubernetes cluster (Minikube, cloud cluster, etc.)
Sufficient permissions to query cluster resources

Instructions

Step 1: Verify Cluster Connectivity

Check if kubectl can connect to the cluster:

# Check cluster info
kubectl cluster-info

# Verify current context
kubectl config current-context

# List all nodes
kubectl get nodes

Expected Results:

Cluster-info shows running control plane
At least one node in Ready state
No connection errors

Step 2: Check Node Resources

Verify sufficient resources are available:

# Get detailed node information
kubectl describe nodes

# Check resource usage (requires metrics-server)
kubectl top nodes

Minimum Requirements:

CPU: >2 cores available
Memory: >4GB available
Disk: >20GB available
Status: All nodes Ready

Step 3: Verify System Components

Ensure essential Kubernetes components are running:

# Check system pods
kubectl get pods -n kube-system

# Check critical components
kubectl get pods -n kube-system | grep -E 'coredns|etcd|kube-apiserver|kube-controller|kube-scheduler'

All pods should be:

Status: Running
Restarts: <3 (some restarts normal during startup)
Ready: All containers ready (e.g., 1/1, 2/2)

Step 4: Validate Storage

Verify storage classes are available:

# List storage classes
kubectl get storageclass

# Check for default storage class
kubectl get storageclass -o json | jq -r '.items[] | select(.metadata.annotations."storageclass.kubernetes.io/is-default-class"=="true") | .metadata.name'

Requirements:

At least one storage class exists
One marked as (default)

Step 5: Test Networking

Validate cluster networking:

# Check kube-proxy
kubectl get pods -n kube-system | grep kube-proxy

# Verify DNS resolution
kubectl run -it --rm debug --image=busybox --restart=Never -- nslookup kubernetes.default

# Check CNI plugin
kubectl get pods -n kube-system | grep -E 'calico|flannel|weave|cilium'

Networking is healthy when:

kube-proxy pods running
DNS lookups succeed
CNI plugin pods running

Validation

Pre-deployment checks passed when:

[ ] kubectl connects successfully
[ ] All nodes are Ready
[ ] System pods running (coredns, etcd, kube-apiserver)
[ ] Default storage class exists
[ ] DNS resolution works
[ ] Resources meet requirements (2 CPU, 4GB RAM)

Troubleshooting

Problem: Cannot connect to cluster

kubectl config get-contexts
kubectl config use-context <correct-context>

# For Minikube
minikube status
minikube start

Problem: Nodes NotReady

kubectl describe node <node-name>

# Restart Minikube with more resources
minikube stop
minikube start --cpus=4 --memory=8192

Best Practices

Always Check First: Run health checks before any deployment
Save Reports: Keep health reports for troubleshooting
Automate: Add to CI/CD pipelines
Monitor Trends: Track resource usage over time
Document Baseline: Know what "healthy" looks like

Agent Skills: Kubernetes Health Check

Skill Files