How Kubernetes Starts a New Pod
The lesson begins with a Deployment replica count changing from two to three. It follows the API object to the Deployment controller, ReplicaSet, scheduler, node agent, and container runtime. A status timeline distinguishes “accepted by the API” from “container started” and “Pod ready for traffic.”
The closing scene shows common stall points such as insufficient resources, image-pull failure, and a failing readiness probe. It also distinguishes control-plane reconciliation from the packet path that reaches a Service endpoint. Companion material: desired-state article, reconciliation trace, and Kubernetes Deployments.
Related articles
Desired State Turns Infrastructure Into a Control Loop
Follow the cloud control plane that compares declared configuration with live resources and works toward the requested state.
Deployments Reconcile Desired State
Kubernetes controllers continuously compare observed cluster state with a declared desired state.
C++ Concurrency: Threads and Atomics
How std::thread, std::mutex, and std::atomic work, and the race conditions and memory ordering issues they prevent.
New lessons by email
Get new articles and notes on the systems behind everyday software.
One technical dispatch per week. No noise.
Not started
Sign in to save your learning progress.