510 posts tagged with "Kubernetes"
Container orchestration, Cluster API, and self-hosted control planes
KEDA's Queue-Depth Autoscaling vs. Render and Railway's Black-Box Scaling
A practical comparison of KEDA's queue-aware Kubernetes autoscaling with Render and Railway's documented CPU, memory, and replica controls, including a worked worker-backlog example.
Kubernetes Metrics API Is Stable in v1.37: The Contract Agents Need Before Autoscaling Apps
Kubernetes v1.37 makes metrics.k8s.io stable, but stable data is not safe automation. Here is the concrete, tenant-scoped contract a self-hosted PaaS should expose before an AI agent changes replicas.
Kubero vs Porter vs Coolify: What Happens When Your Self-Hosted Heroku Needs a Second Server?
Kubero, Porter, and Coolify all bring Heroku-like deployment to infrastructure you control. This comparison shows where app deployment ends and machine provisioning, replacement, traffic routing, and fleet reconciliation begin.
8% CPU, 69% Overprovisioning: A 30-Minute Kubernetes Fleet Audit
Measure allocatable, requested, and used resources in 30 minutes, then test whether a self-hosted Kubernetes fleet can safely remove a node without sacrificing N-1 resilience.
Before AI Agents Operate Your Platform: The Catalog Hygiene Contract Your IDP Needs
Use a field-level catalog contract, a worked deploy example, and staged authority controls to let AI agents safely operate a self-hosted developer platform.
CNCF's 276–327-Patch Maintenance Bill for an Internal Platform
A 14-project Kubernetes platform can face 276–327 patches, 43–52 minor upgrades, and 2–5 major upgrades a year. This worked model turns that queue into 187–1,063 engineer-hours and shows how part-time operators can contain it.
Coolify Critical CVEs: The PaaS Isolation Lesson
Critical Coolify advisories show how authorization failures and deployment input can reach privileged infrastructure—and which layered controls can narrow the blast radius.
Crossplane vs. Cluster API: Draw the Boundary Before AI Draws It for You
Crossplane and Cluster API use Kubernetes reconciliation, but they own different things. Map the boundary between cluster lifecycle, external services, and tenant APIs before exposing infrastructure to AI agents.
CI Runner Pricing in 2026: Depot, Blacksmith, and the Cost of Reserving Your Own Build Capacity
A reproducible CI-runner cost model across 2, 8, and 20 concurrent jobs: current Depot and Blacksmith pricing beside the fixed cost of reserving self-hosted build nodes.