1512 posts tagged with "PaaS"
Platform-as-a-Service tooling, economics, and self-hosted alternatives
Dokku vs Dokploy vs Kubero: Three Generations of Self-Hosted Heroku Alternatives
Dokku, Dokploy, and Kubero are three generations of the same idea: Heroku-style deploys on infrastructure you own. A head-to-head on deploy flow, scaling, HA, multi-node reality, and upgrade pain, plus where declarative machine provisioning fits next.
Dokploy in 2026: How Close Is Single-Server PaaS to Cluster API?
Dokploy spent 2026 adding remote servers, previews, external secrets, and agent tooling. This inventory sorts which additions actually raise the single-box ceiling from dashboard polish, and names the machine-layer line that still needs declarative provisioning.
Stop Letting GPU Requests Pend: Kubernetes 1.36's DRA Prioritized Alternatives
Exact-match GPU requests leave pods pending while other card types sit idle. Kubernetes 1.36's stable DRA prioritized lists let one claim say H100, A100, or T4 — here is the claim-template design, the scoring behavior, and four gotchas for mixed-card fleets.
Fly.io's Seven Incidents in Nine Days: What Reading a Status History as a Migration Signal Actually Looks Like
Fly.io logged seven status-page incidents in nine days across builders, edge network, metrics, certs, and datacenter power. A tally of what broke, whether it is a pattern or a bad week, and a five-check method for reading any vendor's status history as a migration input.
Fly.io's Trial-Credit Runway: What Month 13 Costs When the $5 Runs Out
A $5 Fly.io credit lasts ten months for an idle side project and eighteen days for an always-on starter. Here is the line-by-line month-13 bill versus a flat-rate Hetzner box, and why a credit with no free tier behind it is a countdown.
formae Learned Kubernetes. Does a Cluster-API Fleet Still Need Another IaC Layer?
formae's May 2026 release added Kubernetes, Helm, and .tfvars support to its agent-based, state-file-free IaC platform. Here is where it fits on a Cluster-API fleet that is already GitOps-native — and where a second reconciler would only fight the first.
Only Two Free Tiers Left: What Render and PythonAnywhere Still Give Away vs. a $5 Hetzner Box
Only Render and PythonAnywhere still offer free-forever compute in September 2026. A line-by-line comparison of what each survivor covers — including Render's 30-day database cliff — against a ~$5/month Hetzner CX22, and which side of the free-vs-owned line a side project should start on.
Shipping a Deploy-Authority MCP Server on Golf: What the Framework Owns and What You Still Build
Golf, a production MCP server framework on FastMCP 4.0, owns auth, telemetry, and transport so you write only agent logic. Here is the exact framework-vs-platform split for an MCP server that deploys to production — plus the credential scoping, deploy API, and audit log no framework builds for you.
Headlamp Ate the Kubernetes Dashboard: What Cluster API Operators Actually Get (and What a Git-Push PaaS Should Still Hide)
The Kubernetes Dashboard is archived and Headlamp is the official successor. What the migration changes for Cluster API operators, what the new CAPI plugin delivers, and where a cluster console belongs in a git-push platform.