Debugging is the systematic investigation of application and cluster behavior to find root causes, restore healthy operation, and limit downtime. In Kubernetes, it distinguishes application, node, control-plane, operating-system, and network failures. section 1.4.5 section 8.0
Use observable evidence to narrow the failure:
- Inspect container output with kubectl or files under /var/logs/containers; use logs to understand application behavior and imminent failures. section 8.4
- Check pod statuses, cluster-wide events, and metrics to correlate symptoms with resource usage and cluster conditions. section 8.4
- Examine kube-system components for control-plane problems, then investigate nodes, kube-proxy, CNI configuration, ports, labels, Services, and Deployments for infrastructure or connectivity faults. section 8.4
- Apply a top-down approach when errors compound: peel back dependent layers until the root problem becomes identifiable. section 8.4
|
Competency
|
Chapter section
|
|---|---|
| Monitor cluster and application resource usage |
8.1 |
| Manage and evaluate container output streams |
8.1 |
| Troubleshoot cluster components |
8.1, 8.2, 8.3 |
| Troubleshoot clusters and nodes |
8.2 |
| Troubleshoot Services and networking |
8.3 |
Map debugging work to operational competencies: resource monitoring, container output, cluster components, nodes, Services, and networking. section 8.0
Practice scenario
Send the SWAPI request starships/9/, compare its fields with a Kubernetes Deployment, list all API resources, and save the output as resources.csv. section 1.4.5
