buoy

Every service goes down eventually. The question is: do you find out from your users, or from your monitoring? buoy is a self-hosted uptime monitoring dashboard that pings your endpoints at configurable intervals and shows you exactly what's happening.

Named after the floating markers that track ocean conditions, buoy does the same for your web services. Add a URL, set a check interval (30 seconds to 5 minutes), and buoy starts pinging. The dashboard shows response times as sparkline charts, calculates uptime percentages, and lights up green or red so you can see status at a glance.

It's deliberately lightweight. Express serves both the API and the frontend. SQLite stores all the check history locally — no external database to configure. The frontend is vanilla HTML/CSS/JS with an ocean-themed color palette (deep blues, teals, coral accents). No build step, no framework overhead, just a clean dashboard that works.

The embeddable SVG badges are a nice touch — drop one in your README to show live status. And the full JSON API means you can build your own integrations or pipe data into other tools.

The main design challenge was keeping everything self-contained. No external databases, no build tools, no frontend frameworks. Just npm install and go. This meant using SQLite for storage (via better-sqlite3 for synchronous queries), Express for both API and static file serving, and vanilla JS for the dashboard.

The monitoring loop needed to handle failures gracefully — timeouts, DNS errors, connection refused — without crashing the whole server. Each check is wrapped in a promise that always resolves, capturing errors as data rather than throwing.

Sometimes the best monitoring tool is the simplest one. Big platforms like Datadog and PagerDuty are great for enterprises, but for a handful of personal projects and services, a single Node.js process with a SQLite database does the job perfectly.

The ocean theme made the design decisions easy. Deep navy background, teal accents for healthy services, coral red for failures — it all maps naturally to the buoy metaphor. When your service is "afloat," the buoy glows green. When it sinks, the buoy goes red. The metaphor does the UX work.