Infra Learning Club shares study notes, cutting-edge technology, and career discussions.
Debugging Kubernetes components is challenging due to version mismatches, service dependencies, and IDE limitations; this guide outlines two practical methods—log‑based analysis with redeployed services and direct IDE debugging—while detailing environment setup, local and remote development workflows, and testing procedures.
This article dissects Kubelet's Pod admission pipeline, explaining how syncLoopIteration gathers pod data, how HandlePodAdditions invokes canAdmitPod, and how six registered admit handlers—Eviction, System Allowlist, Resource Allocation, Predicate, AppArmor, and Shutdown—evaluate each pod with concrete code examples and decision logic.
The article outlines four ways to extend the Kubernetes scheduler—Scheduler Extender, Scheduler Framework, deploying multiple schedulers, and a WebAssembly‑based plugin model—detailing their mechanisms, configuration steps, advantages, and drawbacks.
The article analyzes Elasticsearch’s shift from Apache 2.0 to a dual ELv2 + SSPL/AGPL licensing model, explains the competitive pressure from AWS, and offers guidance on choosing appropriate open‑source licenses for different types of foundational software.
This article dissects Kubelet's ImageManager and ImageGCManager, detailing their interfaces, implementation structs, pull‑policy handling, parallel image pulling, GC policy thresholds, and the step‑by‑step execution flow that governs image lifecycle on a node.
The article surveys open‑source GPU sharing and virtualization approaches for AI workloads, comparing soft isolation, CUDA‑level isolation, NVIDIA MPS, driver‑level isolation, GPU pooling and deep‑learning memory sharing, and highlights their architectures, isolation guarantees, and performance trade‑offs.
The article outlines a comprehensive, bottom‑up learning path for containers, covering Linux container fundamentals, images, runtimes, managers, orchestrators, and non‑Linux implementations, while clarifying common misconceptions, explaining how runtimes like runc work, and comparing Docker, containerd, and Kubernetes.
This article explains how Kubernetes uses the device‑plugin framework to extend resources beyond CPU and memory, details the kubelet registration and allocation workflow, and walks through the relevant source code in pkg/kubelet/cm/devicemanager that builds the OCI spec.
The article dissects Kubelet’s VolumeManager, detailing its asynchronous loops, the VolumeManager interface, how it is started from Kubelet.Run, the handling of Attach/Mount and Unmount operations during pod sync, the internal struct fields, and the plugin initialization process that together manage the full lifecycle of pod volumes.
This article dissects Kubelet's core data structures—podManager, podWorkers, and workQueue—explaining how they store and synchronize regular pods, static pods, and mirror pods, and detailing the lifecycle steps for pod creation, deletion, and cleanup.
