Marcus Lee

Spent all morning debugging a Kubernetes issue. Turns out someone set replicas to 0. That someone was me. Last Friday.

Terraform plan: 0 to add, 0 to change, 0 to destroy. Most satisfying output in all of DevOps.

Just realized our Prometheus retention was set to 15 days. We've been flying blind for every incident older than two weeks.

Grafana + Prometheus + Loki = the observability holy trinity. Open source and genuinely good.

Kubernetes tip: if your pod keeps crashing, check resource limits before blaming the code. Learned this the hard way today.