Lantern
A lightweight, self-hosted service health dashboard written in Go that auto-discovers Docker containers and monitors uptime via heartbeats and active HTTP/TCP/ICMP checks.
🔗 Visit Lantern
Description
If you run a handful of your own servers or Docker containers, you eventually want one page that just says "is everything up?" Lantern is a small, self-hosted dashboard that watches your services for you — either by pinging them or by the services checking in themselves — and lights up red the moment one goes down.
Lantern is a single Go binary (SQLite storage, no CGO) that auto-discovers Docker containers with zero config, accepts push-based heartbeats, and runs active HTTP/TCP/ICMP checks with regex or JSON-path validation on responses. It adds production touches many weekend-project status pages skip: flap dampening and quiet hours to cut alert noise, TLS certificate expiry tracking, role-based access (owner/admin/viewer), audit logging, and notifications to Discord, Telegram, Gotify or webhooks. MIT licensed, stable since v0.60.0.
💬 Our review
The short version: Lantern is the self-hosted alternative to "just use a SaaS uptime monitor" for people who'd rather own their status data and don't mind running one extra Go binary.
Against Uptime Kuma, the most common self-hosted comparison, Lantern's Docker auto-discovery and single-binary/SQLite deployment make setup a bit lighter, and its flap dampening plus quiet hours are aimed squarely at cutting 3am false alarms. Against Datadog or New Relic, it obviously doesn't do APM or tracing — it's a focused uptime/status tool, not a full observability suite. Free and MIT-licensed, it's a solid pick if you're self-hosting a handful of services and want a status page without a subscription; skip it if you need distributed multi-region monitoring or already have Datadog covering this.
💰 Pricing
📊 Global score
🤖 AI-enriched data
Gratuit, licence MIT
Pros
Auto-découverte Docker sans configuration
Déploiement en un seul binaire Go + SQLite
Anti-bruit intégré (flap dampening, quiet hours) et suivi d'expiration TLS
Cons
Modèle mono-serveur (SQLite), pas de clustering distribué
Communauté encore modeste face aux suites enterprise