StatOSS

An open source status page

The core of StatOSS is MIT licensed: the checker and the public page in one container, configured with a YAML file. Run it yourself, or use the hosted service.

northwind.statoss.comPublic page

Northwind is up.

All 3 monitors are responding, 2 failed checks in the last 24 hours.

Website

Up for 23 h 58 min100% passed

1,438 checks, median 137 ms, all passed

95% under 199 ms

API

Up for 7 h 6 min99.86% passed

1,438 checks, median 204 ms, 2 timeouts

95% under 299 ms

  • Timed out2 min, 2 checks

Docs

Up for 23 h 58 min100% passed

1,438 checks, median 94 ms, all passed

95% under 136 ms

The page the standalone draws, with made-up data: the same headline and strips as the hosted service, from the same code.

What the core holds

statoss-standalone is the single-tenant version: HTTP, TCP, DNS, ping, certificate, domain and heartbeat monitors; the strip that shows every check with its response time and every failure with its reason; the public page with incidents, maintenance windows that can repeat, components and the vendors you depend on; alerts to email, Slack, Discord, PagerDuty, Opsgenie, ntfy and a signed webhook; status.json, the badge, the feeds and a maintenance calendar. It is TypeScript, keeps its data in one SQLite file, and runs as one container. An Uptime Kuma database imports with one command. The licence is MIT, which means you can run it, change it, and ship it inside something else.

A live copy runs at statoss.com/standalone-demo.

Why open source matters here

A status page is what you point customers at when everything else is broken, so it should not depend on a vendor you cannot leave. With the core in the open, the page from the hosted service is the same page you can run yourself from the README, and the hosted account's export holds every check. The rule that decides something is down is in the code too: two failed checks in a row, one success to recover.

What the hosted service adds

The hosted StatOSS is the same checks and the same page, plus what a single container cannot be: accounts and a dashboard, a second server on another network that confirms every failure before it counts, six more check types (TCP, DNS, ping, certificate, domain expiry, heartbeat), Microsoft Teams, Telegram and Pushover alerts, email subscribers, a monthly uptime report, your own domain with the certificate handled, the REST API and the MCP endpoints, and the server, backups and failover run by us. It starts free; Hobby is $4 a month and Pro $15. The pricing page has the table.

Other self-hosted options

Uptime Kuma is the most popular self-hosted monitor, with dozens of check types and a status page; the comparison goes through where it fits better. Gatus (Apache 2.0) and openstatus (AGPL-3.0) both check and publish a page, and both have a hosted service; the comparisons with Gatus and openstatus go through them. Cachet and Upptime are the other names people search for. Cachet is a status page without monitoring that you update through its API, under Alt Three Services' own licence (commercial use and changes allowed, no redistribution as a standalone product), with version 3 unreleased as of July 2026. Upptime runs checks every five minutes from GitHub Actions and publishes a static page from the repository; its last release was in October 2020. StatOSS is MIT licensed and runs as a hosted service on the same code.