k8s-query

Query Kubernetes Resources

Check status of Kubernetes pods, deployments, services, and events.

When to Use

• Checking if pods are running
• Viewing pod/deployment status
• Investigating pod failures or restarts
• Checking resource events and logs

Pre-flight Checks

Authentication and Context

# Check kubectl context
kubectl config current-context || {
  echo "No kubectl context. Run: kubectl config use-context <context>"
  exit 1
}

# Show current context and namespace
CURRENT_CONTEXT=$(kubectl config current-context)
CURRENT_NAMESPACE=$(kubectl config view --minify -o jsonpath='{..namespace}' || echo 'default')
echo "K8s Context: $CURRENT_CONTEXT"
echo "Namespace: $CURRENT_NAMESPACE"

# If context suggests different environment, prompt to switch
# Example: User mentions "production namespace" but current context is "staging-cluster"
# Detect from query context and prompt:
# echo "Query mentions 'production' but current context is '$CURRENT_CONTEXT'"
# echo "Switch to production context? Run: kubectl config use-context prod-cluster"
# read -p "Continue with current context? (y/n) " -n 1 -r

Query Strategy

Follow this approach for effective Kubernetes debugging:

1. Check pod status first: kubectl get pods -n <namespace>
2. Check events if pod failing: kubectl describe pod <pod-name> -n <namespace> (events at bottom)
3. Check logs if pod running: kubectl logs <pod-name> -n <namespace>
4. Check previous logs if crashed: kubectl logs <pod-name> --previous -n <namespace>
5. Correlate with GCP logs: Save K8s events/logs to /tmp, check GCP for service logs

Example Workflow

# Step 1: Check pod status
kubectl get pods -n production -l app=api-gateway

# Step 2: If pod is failing, describe for events
kubectl describe pod api-gateway-abc123 -n production
# Look at Events section at bottom for errors

# Step 3: If pod running but misbehaving, check logs
kubectl logs api-gateway-abc123 -n production --tail=200

# Step 4: If pod crashed/restarted, check previous logs
kubectl logs api-gateway-abc123 --previous -n production

# Step 5: Check namespace events for broader context
kubectl get events -n production --sort-by='.lastTimestamp' | tail -20