Kube Architect

This case study shows how a JVM DNS cache pinned one service to dead AWS Network Load Balancer IPs during migration, causing an outage fixed with a 30-second TTL. More: https://ku.bz/SvLNyzRrh

This tutorial walks through building a Crossplane platform API on a local kind cluster, modelling an InfluxDB provider with XRDs and compositions, and debugging the composition when it fails. More: https://ku.bz/msSlqDZKr

This article explains how the Kuadrant Console Plugin turns raw policy YAML into guided forms inside the OpenShift console, and how the form and YAML views stay in sync in both directions. More: https://ku.bz/bZqvJv5VL

This case study shares how a team upgraded a production OpenShift cluster on IBM Cloud with Portworx replication in place, and which compatibility and worker-replacement checks kept it from turning into an outage. More: https://ku.bz/fZr2FXszm

This article argues that uptime, latency and error rate only tell you a system is alive, not correct, and shows which behaviour-level signals and baselines belong in dashboards and alerts instead. More: https://ku.bz/v09SHsxGG

This case study shows how a team gives every feature branch its own staging environment on Kubernetes, generating manifests in CI and letting an Argo CD ApplicationSet create and delete them. More: https://ku.bz/FLNLH90vp

This article walks through a real-time object detection pipeline where Kafka absorbs bursty camera traffic, YOLOv8 workers autoscale on Kubernetes, and MLflow tracks every model version. More: https://ku.bz/vKPdn6ST4

This case study shows how Zalando moved over one million requests per second to in-process load balancing, kept caches warm during scaling, and saved more than $1,000 a day. More: https://ku.bz/p9Lpdy2Hh

Frédéric Gaudet, Senior SRE @ BlaBlaCar, explains the critical relationship between Kubernetes deployment strategies and Istio's warm-up feature effectiveness. He reveals that without proper rollout configurations, warm-up becomes completely ineffective because all pods get terminated simultaneously, breaking traffic share calculations. The solution requires implementing rolling deployment strategies with either maxUnavailable or maxSurge parameters. Watch the full episode: https://ku.bz/grxcypt9j

This week on Learn Kubernetes Weekly 206: 🌐 DNS Is Weird Inside Kubernetes on AWS ⏳ TIL: Kubernetes ConfigMap Volume Mounts Are Eventually Consistent — Not Immediately Consistent 🚨 CVE-2025-69902: OS Command Injection in kubectl-mcp-tool 🐘 Building the Foundation for PostgreSQL Lifecycle Management in a DBaaS on Harvester 💰 We Cut Our Cassandra Bill From $30K to $10K a Month by Self-Hosting on Kubernetes Read it now: https://kube.today/issues/206 ⭐️ This issue is brought to you by Teleport — stop sharing credentials and give every engineer, service, and AI agent a login that expires automatically https://ku.bz/2tJxlTyDy

LearnKube Day includes ten five-minute lightning talks from Kubernetes practitioners sharing practical lessons with the community. Graziano Casto, Developer Advocate at Akuity, is one of the featured lightning-talk speakers at LearnKube Day Salt Lake City 2026. The event brings together, lightning talks, YAML Games, and a community reception with a live DJ(our own Amin Astaneh). Learn more: https://learnkube.com/learnkube-day-salt-lake-city-2026

This case study shows how an insurance platform scaled from 50 to 200,000 insured objects per policy with parallel batch processing, dynamic workers, Kubernetes autoscaling, database scaling, and caching. More: https://ku.bz/jybsg5c_0

This tutorial shows how to run a multi-tenant Spark platform on Kubernetes with Spark Operator, Volcano, Helm, Airflow, shared images, workload limits, and monitoring. More: https://ku.bz/-GBQPG_S6

CPU and memory usage need context before you change Kubernetes requests and limits. Throttling, pressure, and runtime events help explain the usage graph. Looking at each replica separately shows differences that a workload average can hide. We are releasing the Kubernetes Resource Review Mixin, an open-source Grafana dashboard with Prometheus recording rules. You can use it to: - See how each Pod's CPU and memory usage compares to its requests and limits. - Inspect CPU throttling, CPU and memory pressure when PSI metrics are available, and OOM and restart activity. - Review replicas and workload revisions separately, with p95, maxima, and resource differences in the tables. The mixin is an extension of our book on rightsizing Kubernetes workloads, The Technical Guide to Kubernetes Rightsizing. Clone it here: https://github.com/learnk8s/resource-review-mixin

This tutorial shows you how to implement service discovery in .NET with Kubernetes DNS, Microsoft.Extensions.ServiceDiscovery, Consul, and Dapr while ordering resilience handlers correctly. More: https://ku.bz/3HKvQvvKR

This article explains why delayed metrics weaken Kubernetes autoscaling and how real-time demand signals, KEDA, and workload-specific scalers create faster, more reliable scaling loops. More: https://ku.bz/-_4ghyZgD

This case study shows how a retail platform manages hundreds of Kustomize overlays with separate branch and kiosk settings, storage choices, and GitOps updates. More: https://ku.bz/1pM9vvzxQ

This case study shows how to shut down 75 AKS services overnight by pausing KEDA, scaling deployments to zero with Helm CronJobs, and letting Argo CD keep everything in sync. More: https://ku.bz/rnDvkmmld

This case study shows how a team cut its Cassandra bill from $30K to $10K a month by running K8ssandra across four EKS regions and moving traffic with no downtime. More: https://ku.bz/tJ6Yy3JQ_

Danyl Novhorodov explores how BEAM runtimes (Erlang, Elixir) and Actor model systems (Orleans, Akka.NET) provide built-in fault tolerance and scalability without requiring external orchestration like Kubernetes. He contrasts Kubernetes' approach of achieving reliability through external mechanisms (probes, sidecars, rescheduling) with Actor models that embed resilience directly into the application layer. The discussion covers how BEAM systems embrace process crashes as normal behavior, using supervisors to automatically restart failed processes, while Actor models provide location transparency and automatic lifecycle management. Watch the full episode: https://ku.bz/BYhFw8RwW