Hey! :wave: A little bit more than a year ago I t...

# feed

[Adam Kobor]

Hey! 👋

A little bit more than a year ago I took my pet project called “Kuvasz Uptime” out of the drawer and gave it a new life by almost completely revamping it. It was more of a fun thing to do, because primarily I just wanted to play around with SSR and kotlinx.html without any pressure. And this is how 2.0.0 was born, which I also shared here if I recall correctly.
The revamp was so successful that it got some attention in the self-hosted and homelab communities, and also among smaller companies who wanted to get rid of their UptimeRobot bills. This success made me continue working on it, so now it stands at 4.3.2, and a lot of interesting features have been built since 2.0.0:

- New monitor types next to HTTP & SSL: push/heartbeat monitors for cron jobs and services you can’t reach from the outside, ICMP ping, TCP (any host:port), and DNS monitors with per-record assertions (A, MX, TXT, SRV, CAA, …), and opt-in drift detection
- Status pages: public or private, brandable pages with uptime history
- Maintenance windows: manual, recurring (cron) or one-off planned downtime that pauses the checks and suppresses false alerts
- Custom webhooks: fully templatable body, headers and even the URL, with a configurable HTTP method — so basically anything with an HTTP endpoint can be notified
- New built-in integrations: Discord, Microsoft Teams (Adaptive Cards), Pushover and Apprise
- Metrics exporters for Prometheus and OpenTelemetry, so it plugs into your existing observability stack
- A built-in MCP server, exposing your monitors, incidents and status pages as tools your AI assistant can call
- OIDC/SSO login (Keycloak, Auth0, Okta, …)
- Much richer HTTP checks: expected status codes, keyword matching in the body (case-sensitive and negatable), response-time thresholds, expected and custom request headers, request bodies, all the usual methods
- Incidents & per-monitor stats: a dedicated incident view with filtering, plus uptime ratio, incident count and total downtime broken down per monitor
- Backup & restore: the whole configuration — monitors, maintenance windows and status pages — can be exported and re-imported as YAML, right from the UI or the API
- Deployment & community bits: an official Helm chart, a dedicated Home Assistant integration (ku-hass), a live demo instance
- Under the hood: it now runs on Kotlin 2.4 and Micronaut 5, still SSR with kotlinx.html + htmx and Alpine.js

It’s still nothing more than a pet project, but if you want a simple and straightforward, yet stable uptime monitor, then give it a try - you won’t be disappointed!
Contributions and feedback are all welcome!

github.com/kuvasz-uptime/kuvasz

[Mohammed Alaa]

Thanks for your great effort, really useful project 💯