1732 posts tagged with "Self-Hosting"
Running your own PaaS and infrastructure on machines you own
Kubernetes 1.37 Takes HPA Scale-to-Zero to Beta: What Idle Finally Costs vs Fly.io, Render, and Railway
HPA scale-to-zero is beta and on by default in Kubernetes 1.37. A duty-cycle idle-bill table pits native minReplicas: 0 against Fly.io, Render, and Railway — plus the HTTP buffering gap to close before idle costs zero.
'Kubernetes Is the PaaS': What the Portability Thesis Gets Right, and the Operations Bill It Skips
Hacker News keeps rediscovering the thesis that Kubernetes is a vendor-neutral PaaS layer. The portability half holds up — workload manifests really do run anywhere — but the operations half never disappeared: upgrades, etcd, networking, and node lifecycle still need an owner.
KYAML Goes Stable in Kubernetes v1.37: What a Canonical YAML Subset Means for the Manifests a Git-Push PaaS Generates on Every Deploy
Kubernetes 1.37 promotes KYAML — a strict, unambiguous YAML subset with a stable kubectl output — to GA. What canonical manifests change for platforms that generate YAML on every deploy, and why compliance belongs at emit time.
MCP's 2026 Roadmap Bets on Stateless Scale: What It Unlocks for a Self-Hosted Deploy Server
MCP's 2026 roadmap puts stateless transport ahead of new features, and the July 2026 spec delivers it: no sessions, no sticky routing, any replica behind an ordinary load balancer. Here is what that unlocks for a self-hosted deploy server — and what is safe to build today versus still landing.
MCP Lock-In Moves Up the Stack: Why Your Infrastructure MCP Server Belongs on Your Own Fleet
MCP passed 10,000 public servers and 97M monthly downloads under foundation governance — but lock-in moved to registries, client defaults, and trust signals. Why teams running deploy-capable infrastructure servers should self-host the server, the catalog, and the registry surface.
MinIO Vanished From Docker Hub: The One-Line Fix and the Migration You Actually Owe Yourself
MinIO deleted minio/minio and minio/mc from Docker Hub in mid-September 2026, breaking every cold docker compose pull. The quay.io repoint restores green builds today — but that image is frozen forever, so here is the honest SeaweedFS vs Garage vs RustFS comparison plus the registry mirror that makes the next deletion a non-event.
Mistral Bought Koyeb: The Tenant Exit Playbook
Koyeb's February 2026 acquisition by Mistral pivoted the platform toward AI inference — this tenant exit playbook rates every Koyeb primitive, from Light Sleep to Serverless Postgres to per-second GPUs, on how cleanly it maps to infrastructure you own, with a move order that keeps each step reversible.
Resize the Queue, Not the Pods: Kubernetes 1.36's Mutable Job Resources for Bin-Packing AI Sandboxes
Kubernetes 1.36 promotes mutable pod resources for suspended Jobs to beta: patch a queued job's CPU, memory, or GPU ask to fit free capacity instead of deleting and recreating it. How the suspend-resize-resume flow works, what it buys bursty AI-sandbox pools on owned nodes, and where it still falls short.
Netcup Vouchers vs Hetzner Post-Hike Prices: A Fleet-Role Placement Guide
Hetzner's June hike repriced dedicated cores to €21.50 while netcup's September vouchers cut RS root servers to about €2.24 per core. A role-by-role placement guide for small EU fleets — control plane, elastic workers, batch — plus an honest verdict on second-provider diversification.