Patterns
Reusable ways to work through engineering problems, from Kubernetes and Linux troubleshooting to incident triage and cloud architecture, with lessons and labs.
- Kubernetes troubleshooting patterns — How to work through a broken cluster: NotReady nodes, crash loops, Services with no endpoints, Pending pods and names that will not resolve.
- Linux troubleshooting patterns — How to work through a misbehaving Linux machine: full disks, services that will not start, permission errors, unreachable services and stuck processes.
- Incident triage patterns — How to work through an incident: judge the alert, run the response, rate severity, read the first failure, contain an attacker and learn from it.
- Cloud architecture patterns — Reusable cloud designs: least privilege, zones and failover, private networks, scaling, tested recovery, caches and queues, timeouts and retries.