Service Debugging and Troubleshooting
You'll learn how to diagnose and fix Service connectivity problems using kubectl and DNS queries to verify that traffic reaches the right Pods.
What Happens When Services Don't Work?
When applications can't connect through a Service, it's like having a phone number that doesn't ring anywhere. The issue could be with the "phone book" (DNS), the "routing" (Service configuration), or the "destinations" (Pod endpoints). Kubernetes gives you tools to check each layer.
Using kubectl to Debug Services
Start by checking if the Service exists and has endpoints:
kubectl get service my-service
kubectl describe service my-service
The describe output shows the Endpoints section—these are the actual Pod IPs the Service routes to. If this list is empty, your Service's selector doesn't match any running Pods.
Check the Endpoints object directly:
kubectl get endpoints my-service
Verify your Pods have the correct labels:
kubectl get pods --show-labels
Compare these labels with your Service's selector from kubectl describe service.
DNS Debugging
Services should be resolvable by DNS. Test from inside a Pod:
kubectl run test-pod --image=busybox -it --rm -- nslookup my-service
This should return the Service's ClusterIP. If DNS fails, the problem might be with CoreDNS or the Service name itself.
For cross-namespace access, use the full DNS name: my-service.my-namespace.svc.cluster.local.
Common Issues
- No endpoints: Label mismatch between Service selector and Pods
- DNS fails: Service name typo or CoreDNS problem
- Endpoints exist but traffic fails: Check Pod readiness and target ports
Key Takeaway: Service debugging follows a checklist—verify the Service exists, confirm it has endpoints matching your Pods' labels, test DNS resolution, and validate port configurations using kubectl describe and DNS queries.