640 posts tagged with "Kubernetes"
Container orchestration, Cluster API, and self-hosted control planes
Kubernetes 1.37 Will Brick Your Fleet on Upgrade Day: ipvs, Static Pods, and cgroup v1
Kubernetes 1.37 lands August 26 with three fleet-breaking changes hiding behind deprecation warnings — ipvs enters a removal clock, static Pods can no longer reference Secrets, and cgroup v1 nodes refuse to start without an override.
Native Histograms Hit Kubernetes 1.37: The ~10x Prometheus Shrink Your Self-Hosted Fleet Pays For in Disk
Classic histograms turn every latency metric into N+2 time series. Native histograms collapse that to one — a ~10x storage win landing in the Kubernetes 1.37 cycle. A worked recompute for a self-hosted PaaS fleet on Hetzner, plus the scrape-config migration that gets there without re-instrumenting.
Temps Wants to Fix Coolify and Dokploy: Why a 99.9% Reliable One-Box PaaS Still Isn't a Fleet
Temps arrived January 2026 promising 99.9% deployment reliability by fixing the bugs that plague Coolify and Dokploy. It's the latest polished one-box self-hoster in a crowded field — and the clearest proof that a more reliable single server is still a single server.
Your Health Checks Are Quietly Killing Scale-to-Zero — And How to Get Idle Actually to Mean Zero on Kubernetes
You configured scale-to-zero but the service never sleeps — health checks and uptime monitors keep waking it. A practical design for genuine idle-to-zero on Kubernetes using KEDA, KubeElasti, and the activator pattern.
Hetzner CX22 vs AX41 in 2026: €4.50 Cloud VM vs €45 Bare Metal for Your Self-Hosted PaaS
A Hetzner CX22 cloud VM costs €4.50 with 20TB included; an AX41 dedicated box costs €45 with 64GB RAM and 1TB NVMe but no API. A head-to-head spec table, worked monthly math for 1/3/10-node fleets, and a checklist for when plain Docker beats Cluster API.
MetalLB vs kube-vip on Bare Metal: The Two-Day Hetzner Outage That Settled a Fleet's Load Balancer Choice
Hetzner's 44-hour Ashburn load balancer outage took down every fleet that trusted a managed L4. A practical comparison of MetalLB and kube-vip on bare metal — their failover behavior, BGP needs, and why a Cluster API fleet on dedicated hardware should run both in different roles.
CAREN Meets Flux: Composable Runtime Hooks for Day-2 Cluster API Without Custom Controllers
CAREN turns Cluster API day-2 operations into composable Runtime SDK hooks — CNI, CCM, CSI, and LoadBalancer GC sequenced through blocking lifecycle callbacks. A concrete walkthrough of wiring CAREN into a CAPH ClusterClass and closing the loop with Flux GitOps.
Chaos-Test Your Cluster API Fleet: Three LitmusChaos Experiments That Prove Self-Healing Before 3 A.M. Does
MachineHealthCheck promises your fleet self-heals — but few teams have watched it happen. A runnable LitmusChaos suite for CAPH/CAPD fleets: kill a kubelet, partition a control plane, fill a disk — and assert remediation fires within nodeStartupTimeout, workloads reschedule, and the fleet converges without a human.
Eleven Minutes, Zero Humans: What Kairos's Self-Healing Upgrades Teach a Bare-Metal PaaS
A CNCF Kubestronaut ran an 11-minute, zero-human Kubernetes upgrade on Kairos's A/B immutable OS — and documented every bug that almost broke it. Here is the pipeline, the two fixes that saved etcd quorum, and how to translate it to a Hetzner + Cluster API fleet.