DevOps Engineering Case Studies

Networking

DNS, TCP, proxies, load balancers and TLS. The failures in this section share a property that makes them hard: the symptom appears at a layer far above the cause, and the metrics of the failing component usually look fine.

Written

None yet.

Planned

Case study Type
DNS resolution failure in a service mesh Troubleshooting
Intermittent 502s from an upstream keepalive mismatch Troubleshooting
TCP connection timeouts under load — backlog, TIME_WAIT, ephemeral ports Troubleshooting
Load balancer health check flapping Troubleshooting
TLS certificate expiry in production Incident
nginx as a reverse proxy: timeouts, buffering, and upstream failover Architecture
  • IPMG — modular IP management and ping-monitoring CLI with parallel network scanning, subnet auto-discovery and scheduled monitoring.

Nothing here yet — content is on the roadmap.