Kai Ning's bilingual infrastructure notebook on Kubernetes, GitOps, observability, AI agents, and production lessons.https://lemonkai.github.io/en-ushttps://lemonkai.github.io/blog/en/gitops-review-checks-before-sync/https://lemonkai.github.io/blog/en/gitops-review-checks-before-sync/A compact pre-sync checklist for GitOps changes so an innocent YAML edit does not become surprising production behavior.Tue, 28 Apr 2026 00:00:00 GMThttps://lemonkai.github.io/blog/en/kai-k8s-road-01-container-to-pod/https://lemonkai.github.io/blog/en/kai-k8s-road-01-container-to-pod/Start with the core Kubernetes abstraction first: why the platform schedules Pods instead of treating each container as the main unit of orchestration.Tue, 28 Apr 2026 00:00:00 GMThttps://lemonkai.github.io/blog/zh/kai-k8s-road-01-container-to-pod/https://lemonkai.github.io/blog/zh/kai-k8s-road-01-container-to-pod/這篇先從 Kubernetes 最核心的抽象開始：為什麼它不是直接管理單個 container，而是用 Pod 當作最小調度單位。Tue, 28 Apr 2026 00:00:00 GMThttps://lemonkai.github.io/blog/zh/probe-misread-restart-loop/https://lemonkai.github.io/blog/zh/probe-misread-restart-loop/有些事故不是程式壞掉，而是健康檢查把原本能活下來的服務判成該重啟。這篇記一個很典型的坑。Tue, 28 Apr 2026 00:00:00 GMThttps://lemonkai.github.io/blog/en/kubernetes-rollout-guardrails/https://lemonkai.github.io/blog/en/kubernetes-rollout-guardrails/A compact baseline for probes, rollout settings, and quick triage before a Kubernetes release becomes an outage.Tue, 28 Apr 2026 00:00:00 GMThttps://lemonkai.github.io/blog/en/agent-runtime-observability/https://lemonkai.github.io/blog/en/agent-runtime-observability/Basic logging is not enough for agent systems; you need traces, tool-level outcomes, and a short path from failure back to prompt or policy.Fri, 24 Apr 2026 00:00:00 GMT