{
  "id": 8493637,
  "title": "VMware Admin's Guide to Kubernetes",
  "url": "https://urgent.news/2026/09/19/vmware-admins-guide-to-kubernetes",
  "topic": "tech",
  "section": "Tech",
  "published": "2026-09-19T16:32:26.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/akilesh_js/vmware-admins-guide-to-kubernetes-2of5"
  },
  "original_language": "en",
  "account": "Kubernetes presents a challenge for VMware administrators because most tutorials are written for developers who know Docker, not administrators accustomed to vSphere. The source material maps VMware concepts to their Kubernetes equivalents, highlighting key differences in the mental models required.\n\nFor example, ESXi hosts become Kubernetes nodes, with each node running containers instead of virtual machines (VMs). VM templates map to container images stored in registries, while virtual switches (vSwitch) and Distributed Virtual Switches (DVS) are defined in YAML configurations. NSX microsegmentation translates to Kubernetes NetworkPolicies, written as code and version-controlled.\n\nSome concepts don't have direct equivalents. For instance, Kubernetes lacks a direct equivalent to VM snapshots. Instead, etcd backup and Velero are used for cluster backup. HA design principles transfer well, with an odd number of control plane nodes ensuring quorum. Resource planning skills from VM right-sizing apply to setting CPU requests/limits and memory requests/limits in pod specifications.\n\nThe biggest mindset shift is understanding that pods are ephemeral, unlike VMs which are static. Application design must accommodate pod death and restart. Debugging in Kubernetes relies on logs, metrics, and traces rather than SSH sessions. Storage in Kubernetes follows pods, unlike the attached storage model in vSphere. Instead of VM snapshots, safety nets are GitOps, Helm rollbacks, and kubectl rollout undo.\n\nThe source provides a five-day action plan for VMware administrators beginning with Kubernetes. Day 1 involves creating a local cluster without VMs. Day 2 involves deploying an nginx web server. Day 3 includes adding monitoring using Helm charts. Day 4 demonstrates NetworkPolicy configuration, similar to NSX microsegmentation. Day 5 involves intentionally breaking and fixing the pod to demonstrate Kubernetes' auto-recovery capabilities.\n\nThe author concludes by emphasizing that a VMware background is a superpower, not a hindrance. Understanding infrastructure deeply gives an edge in Kubernetes, as the thinking remains similar even as tools change. The hardest part is moving beyond the GUI.",
  "summary": "The Problem Nobody Talks About When I first approached Kubernetes, every tutorial assumed I was a developer coming from Docker. Nobody wrote for someone who knew vSphere inside-out — who understood DRS, vMotion, VSAN, NSX, and had opinions about datastore sizing. The Mental Model Map | VMware vSphere | Kubernetes Equivalent | Key Difference | |---|---|---| | ESXi Host | Node | Nodes run…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}