694 posts tagged with "Self-Hosting"
Running your own PaaS and infrastructure on machines you own
Buildpack Rebasing: Patching a Base-Image CVE Across Every Running Container in Seconds
A base-image CVE patched at the platform layer takes one command and seconds. The same fix on a Dockerfile-first platform means hundreds of tenants each rebuilding, pushing, and redeploying on their own schedule — here's the actual cost gap, with real numbers.
Buildpacks RFC 0131: What Build Observability Actually Requires Beyond a Streaming Log
RFC 0131 replaces build scrollback with queryable OpenTelemetry spans. Here's exactly what it specifies, where implementation actually stands, and what a Cluster API-managed kpack pipeline would need to expose the same data before the spec ships.
CircleCI Built a Second CI Pipeline Just for AI Agents — A Git-Push PaaS Already Has One
CircleCI built a whole second CI system — a microVM fleet, a CLI, three separate agent hook integrations — just to get test feedback back to an AI agent in under 60 seconds. A platform that already rebuilds on every git push was one exposed entrypoint away from the same thing.
Claude Code's Agents Now Push and Open PRs Unattended — Where Should Your Deploy Trigger Actually Bind?
Claude Code v2.1.198 shipped background agents that commit, push, and open a draft PR with no human in the loop. Here's the concrete trigger-point table for where a deploy pipeline should actually bind when the pusher might be an agent, plus the four-item audit checklist a self-hosted deploy-from-chat MCP server needs first.
Claude Code's Dynamic Workflows Cap Concurrency at 16, Not Hundreds: What a Real Subagent Burst Does to a Deploy MCP Server
Claude Code's Dynamic Workflows can fan out 1,000 subagents per run, but the runtime caps live concurrency at 16. Here's what that actual burst shape requires from an MCP server that exposes deploy and rollback.
Cloud Repatriation Hit 80%: What GEICO and 37signals' Numbers Actually Mean for a 10-Person Team
GEICO and 37signals both cut infrastructure costs 40-60% by repatriating off the cloud — a worked breakdown of whether that math holds up for a team a fraction of their size, not just the enterprise-scale headline.
Cloudflare's Containers Now Talk by Hostname — Only If You Hand-Wire Every Route
Cloudflare's Containers GA pitches hostname-based routing as a networking upgrade. Reading the actual outboundByHost mechanism shows it's a proxy table you write one route at a time — nothing like what a ClusterIP Service already gives a Cluster-API PaaS for free.
CloudNativePG 1.30's DatabaseRole CRD and Lease Election: What Changes for a Self-Hosted PaaS's Postgres Layer
CloudNativePG 1.30 ships a declarative DatabaseRole CRD and a Lease-based primary election, plus two CVE fixes, right as the 1.28.x line hits end of life — here's what that concretely changes for a PaaS running CNPG as its own managed-Postgres engine.
Cluster API Provider OpenStack Ships v1beta2 Behind a New ORC Dependency: What Declarative Fleet Lifecycle Looks Like on Private OpenStack Instead of Hetzner
CAPO's new v1beta2 API ships behind a hard dependency on openstack-resource-controller — here's what continuous reconciliation buys a Cluster API fleet on private OpenStack that Terraform-on-OpenStack can't, and why the model was never Hetzner-specific to begin with.