1732 posts tagged with "Self-Hosting"
Running your own PaaS and infrastructure on machines you own
Daytona Went Closed-Source: What 'Self-Hostable' Means Now That the Control Plane Is Private
Daytona moved its core to a private codebase in June 2026, freezing the public repo at v0.190.0. A before/after accounting of what self-hosting still covers, a five-question checklist for any vendor's claim, and how E2B, Modal, Beam, Microsandbox, and the Nightona fork score.
Daytona's $3.95/hr H100 vs Owning the Card: The Breakeven Math for GPU Agent Sandboxes
Daytona rents H100 sandbox hours for $3.95 while an RTX 5090 costs $4,700 on the street. A worked rent-vs-own comparison across three usage levels pinpoints the ~50-hour monthly breakeven — plus the Docker-default isolation catch.
Dokploy Moved Its Enterprise Features Behind a New License — Here's Exactly What Changed
Dokploy's January 2026 restructure kept its core Apache 2.0 but put SSO, RBAC, audit logs, and white-labeling behind a source-available license. We map exactly what moved, who it affects, and the five checks to run against any self-hosted PaaS license.
Dragonfly Without the Database: Stop Shipping the Same Gigabytes to Every Node
A 10-node fleet pulling a 1.2 GB image downloads 12 GB per deploy; P2P distribution serves it once. We price the duplicate-byte tax across fleet sizes, unpack Dragonfly's no-database lightweight mode, and compare it with Spegel, Kraken, and a plain registry mirror.
Sandbox Price Parity Is Real, but It Isn't the Whole Bill: What E2B, Daytona, and Modal Charge per Agent Run
E2B and Daytona charge identical per-second sandbox rates while Modal meters idle time at zero. A worked per-run and monthly cost comparison, the billing-rule and isolation fine print, and the breakeven math for running agent sandboxes on a fleet you already own.
Fly.io Added Two Billing Lines in 60 Days: What They Do to a Three-Service Bill
Fly.io started billing volume snapshots in January 2026 and Managed Postgres cross-region traffic in February. A worked before/after bill for a three-service HA stack, the retention and region levers that move it, and the same stack priced on Hetzner hardware.
GPU Autoscaling on Kubernetes With KEDA: Scaling on the One Metric HPA Can't See
HPA can't see GPU utilization — nvidia.com/gpu is an integer device count, so a vLLM pod at 8% CPU and a saturated GPU never scales. A worked KEDA pattern: dcgm-exporter to Prometheus to ScaledObject YAML, a minimal external scaler serving utilization plus queue depth, the scale-to-zero cost math, and where KEDA stops and DRA begins.
Headlamp's Cluster API Plugin: The Free Ops Console Your PaaS Team Doesn't Have to Build
Kubernetes retired its Dashboard in January 2026 and Headlamp's June 2026 Cluster API plugin gives small platform teams a free, upstream-maintained fleet-ops console — Machine health, scaling, and ownership maps included. What to adopt, what remains alpha, and why tenants still need their own dashboard.
Two Dashboard Plugins Exposed the Scheduler Gap Every Git-Push PaaS Ignores Until the First GPU Job
Headlamp shipped dashboard plugins for Volcano and Kubeflow eighteen days apart — Job/Queue/PodGroup map views and API-server-native ML panels. A concrete scheduling-gap table, the three primitives a git-push PaaS lacks, and the exact tripwire for adopting Volcano queues.