← Topics

Kubernetes: start here.

This is the Kubernetes material on the site in the order I would read it if I had to run a cluster this year: the concepts first, then the decisions, then the errors you will meet in the first month.

Kept current

Tool

Kubernetes version support checker

Pick a Kubernetes version and see how long it is still supported upstream and on EKS, GKE and AKS.

Open →Tracker · updated Nov 12, 2026

Kubernetes Release and Deprecation Calendar

A living tracker of supported Kubernetes versions, latest patches and end of life dates, EKS, GKE and AKS support windows, and the removals and deprecations that need action before an upgrade. Kept current, every row sourced.

Open →

Read in this order

  1. Kubernetes Explained (2026): Pods, Deployments, Services for Beginners

    Kubernetes explained for beginners on v1.37: the control plane and nodes, Pods, Deployments, ReplicaSets, Services, Gateway API vs a frozen Ingress, ConfigMaps, Secrets and namespaces, what really happens after kubectl apply, and what you can safely ignore.

  2. Is Kubernetes Still Worth It in 2026? Costs and Alternatives

    Is Kubernetes worth it for a small team in 2026? What it really costs on EKS, GKE and AKS, when ECS, Cloud Run or Nomad fit better, and how to decide.

  3. Kubernetes Requests and Limits Explained: 8% CPU Used, Resize GA in 1.35

    Kubernetes requests and limits in 2026: how QoS classes and oom_score_adj decide who dies, why CPU limits cause CFS throttling, what cgroup v2 changed for OOM kills, in-place resize GA in 1.35, VPA InPlaceOrRecreate, and why clusters still use 8% of their CPU.

  4. Kubernetes Networking Explained (2026): CNI, Services, kube-proxy in 1.37

    How Kubernetes networking works in 2026: the flat pod network model, what a CNI plugin does (Cilium 1.20, Calico 3.33, Flannel 0.28), ClusterIP vs NodePort vs LoadBalancer, EndpointSlices, cluster DNS, and why kube-proxy is moving from iptables and IPVS to nftables or eBPF.

  5. Kubernetes Autoscaling: HPA vs VPA vs KEDA vs Karpenter (2026)

    Kubernetes autoscaling for 2026: when to use HPA, VPA, KEDA, Cluster Autoscaler or Karpenter, what changed in 1.37, and the mistakes that cost money.

  6. Kubernetes Security in 2026: First Attack in 18 Minutes, and What Stops It

    How Kubernetes clusters really get attacked in 2026: anonymous API access and RBAC abuse, cryptomining, exposed dashboards and Helm charts, IngressNightmare, plus Red Hat and Wiz survey numbers, the CIS benchmark, Pod Security Standards, user namespaces GA and image signing.

  7. Ingress NGINX Is Retired: What Replaces It and How to Choose in 2026

    Ingress NGINX stopped receiving fixes in March 2026 after seven CVEs in its final seven weeks. What retirement means, why IngressNightmare mattered, where Gateway API stands at v1.6, and how to pick between Envoy Gateway, Cilium, Istio, NGINX Gateway Fabric, Traefik and kgateway.

  8. EKS vs GKE vs AKS in 2026: Same $0.10 Fee, Very Different Clusters

    EKS vs GKE vs AKS compared for October 2026: all three charge $0.10 per cluster-hour and about $0.50 more for long version support, but free tiers, Auto Mode vs Autopilot vs AKS Automatic, CNI, pod identity, GPUs and forced upgrades differ. Sourced tables and a pick-by-team guide.

Guides and fixes

Latest on Kubernetes

Nov 13, 2026

How to Lock Down an SGLang or vLLM Inference Server (2026)

Step-by-step hardening for self-hosted SGLang and vLLM after the NEAR AI breach: bind to loopback, set separate API and admin keys, allowlist endpoints at a proxy, isolate NCCL traffic, and the auth table I generated from SGLang 0.5.23's own code.

11 minNov 13, 2026

Anthropic Lets You Self-Host Claude Managed Agents, in Private Beta

On 12 November 2026 Anthropic opened a private beta of self-hosted Claude Managed Agents: the API server, agent loop, session history and workers run on your infrastructure, and only model calls leave. What changes for platform teams.

6 minNov 10, 2026

How to Run Kafka on Kubernetes with Strimzi 1.3: KRaft Setup (2026)

Step-by-step Strimzi 1.3.1 guide: install the operator with Helm, run a KRaft Kafka 4.3.2 cluster with KafkaNodePools, manage topics and users as CRDs, connect over mTLS, add Prometheus alerts, upgrade safely and fix under-replicated partitions and consumer lag.

13 minNov 10, 2026

Spot Instances Explained (2026): 2 Minutes vs 30 Seconds, Up to 91% Off

How AWS Spot, Google Cloud Spot VMs and Azure Spot VMs work in 2026: discounts of up to 90 and 91%, 2-minute vs 30-second notices, real interruption buckets from the Spot Instance Advisor, price-capacity-optimized allocation, which workloads fit, and AI training on spot.

13 minNov 10, 2026

How to Self-Host SigNoz with OpenTelemetry: Docker or Helm (2026)

Step-by-step SigNoz v0.149.0 self-hosting with Foundry on Docker or the 0.149.0 Helm chart, an OpenTelemetry Collector 0.165.0 agent config, Python auto-instrumentation, log-trace correlation, retention, alerts, and a no-data troubleshooting table.

12 minNov 10, 2026

How to Deploy Ray Serve on Kubernetes with KubeRay 1.7 (2026)

Step-by-step guide to deploying Ray Serve on Kubernetes with KubeRay 1.7.2 and Ray 2.60.0: install the operator with Helm, write a batched Serve app, define a RayService, add autoscaling and GPU workers, run zero-downtime upgrades, and fix common errors.

12 minNov 10, 2026

Model Serving Frameworks 2026: Triton vs BentoML vs Ray Serve vs KServe

Model serving frameworks compared for 2026: Dynamo-Triton 2.74, BentoML 1.4.40 (now owned by Qualcomm via Modular), Ray Serve 2.60, KServe 0.21 and vLLM, with batching, multi-model, autoscaling, TorchServe's archived status, and when each one fits.

12 minNov 10, 2026

Airflow vs Kubeflow vs Prefect vs Dagster (2026): 6 Orchestrators Compared

Airflow 3.4, Kubeflow Pipelines 2.18, Prefect 3.9, Dagster 1.13, Flyte 2 and Metaflow compared for ML and data pipelines in 2026: assets vs tasks, Kubernetes fit, managed prices, and which team should pick which.

11 min
All Kubernetes posts →Search Kubernetes