1732 posts tagged with "Self-Hosting"
Running your own PaaS and infrastructure on machines you own
Kubernetes 1.37 Deletes Deprecated Kubelet Flags: The Pre-Upgrade Audit That Keeps Your Nodes Joining
Kubernetes 1.37 turns deprecated kubelet flags into fatal unknown-flag errors that stop nodes from joining. The exact failure signature, which flags are in the cleanup wave, and a five-step audit for Cluster API fleets to find every stale flag before upgrade day.
KWOK Before Scale: Load-Test Your Cluster API Controllers Against 200 Fake Nodes Before Your Tenants Do
SNCF proved Cluster API holds at 200 clusters. KWOK lets you prove your own controllers hold as object counts grow — a four-rung ramp from 50 to 400 fake nodes on a laptop, with pass/fail thresholds and the informer-cache checklist the top rung exists to expose.
Let's Encrypt Did the Math for You: What ARI Renewal Exemptions Actually Buy a Self-Hosted PaaS at Tenant Scale
Let's Encrypt caps per-subdomain certificates at 50 per domain a week, but renewals coordinated through ARI are exempt from every rate limit. We work out the onboarding and renewal math for a self-hosted PaaS and what enabling ARI in cert-manager takes.
MCP Went Stateless: What Killing the Session Means for an Endpoint That Deploys Apps From Chat
MCP's 2026-07-28 spec removes protocol-level sessions and the Mcp-Session-Id header entirely. The before/after deployment table — sticky ALB plus session store versus plain round-robin — plus the five things stateless does not fix and what a deploy-from-chat endpoint must change to survive agent load.
Michelin Deleted Its Second CNI: What 70-Plus Clusters With Zero Network Visibility Teach a Self-Hosted PaaS About Day One
Michelin ran 70-plus Kubernetes clusters on two CNIs with no network visibility, then consolidated onto Cilium in two months with zero outages. What the blindness cost, how the migration worked, and a five-item checklist for getting the CNI decision right on day one.
Northflank Killed Its microVM Init Container: What One Deleted Startup Step Reveals About PaaS Isolation Overhead
Northflank quietly removed the default init container from its microVM secure runtime. A concrete look at what that per-pod isolation scaffolding cost, and why a fleet that owns its machines pays for isolation once per node pool instead of on every pod start.
OpenTofu Is the Majority Terraform Engine on One Major Platform: What the 63% Tipping Point Means for Your Fleet's IaC Layer
OpenTofu now runs 63% of workloads on a major IaC platform and powers 72% of new workspaces. We scope the numbers honestly, compare where the fork leads and lags Terraform, cite Fidelity's 50,000-state-file migration, and set the default for a Cluster-API fleet's provisioning layer.
Porter's $300 Rule: Where Heroku and Render Bills Tip Over Into Kubernetes on Hardware You Own
Porter told customers spending under $300/month to stay on Heroku. Re-priced against a small HA Kubernetes fleet on owned Hetzner hardware, the real crossover lands lower — around $150-250 once staging and ops time enter the math.
Every Pull Request Gets Its Own URL: The Real Recipe and Cost Math for Preview Environments on Self-Hosted Kubernetes
A concrete recipe for per-PR preview URLs on machines you own — namespace-per-PR lifecycle, one wildcard certificate, three database options with honest prices, and the scheduling math at 5, 20, and 50 open pull requests.