Features Comparison Pricing Blog Docs Français GitHub
New monitoring generation · Docker & Kubernetes

Your entire stack.
One single container.

Containers, endpoints, SSL certificates, cron jobs, system resources, updates — monitored in seconds. Zero configuration.

maintenant-hero
~17 MB
Idle RAM
0
External dependency
30s
To be operational
2
Runtimes (Docker + K8s)

You're juggling 5 tools.
Or worse, you have nothing.

Uptime Kuma for HTTP, Healthchecks for crons, Grafana+Prometheus for metrics, an SSL checker, and bash scripts. None of them see the full picture.

Before — fragmented monitoring

  • Uptime Kuma — HTTP checks only
  • Healthchecks.io — heartbeats only
  • Grafana + Prometheus + cAdvisor — 3 containers
  • Separate SSL checker
  • No unified view
  • 5 dashboards, 5 configs, 5 maintenances

After — Maintenant

  • Auto-discovered containers (Docker + K8s)
  • HTTP/TCP endpoints monitored
  • Cron jobs and heartbeats tracked
  • SSL certificates validated automatically
  • CPU, RAM, network, disk metrics
  • One container. One dashboard. Everything's here.

Everything you need.
Nothing more.

Containers

Docker and Kubernetes auto-discovery. States, health checks, restart loops, logs. Compose projects grouped automatically.

HTTP/TCP Endpoints

Active probing with latency, status codes, content assertions. Configure via Docker labels — zero YAML files.

Heartbeats / Cron

Unique ping URL per job. Track durations, exit codes, and get alerted when a cron doesn't run.

SSL/TLS Certificates

Automatic detection on your HTTPS endpoints. Expiration alerts at 30, 14, 7, 3, 1 day. Full chain validation.

System Resources

CPU, RAM, network and disk I/O — per container and per host. Real-time charts with configurable threshold alerts.

Update Intelligence

Detects available updates via OCI scan. Compare digests across all your containers. Stop blind docker pull.

Unified Alerts

Webhook and Discord built-in. Pro Slack, Teams, Email, escalation, routing, and custom templates.

Status Page

Real-time public page for your users. Component groups, live SSE updates. Pro Incident management, maintenance windows, subscriber notifications.

REST API + MCP

Full API under /api/v1/. Real-time SSE streams. Built-in MCP server to query your infra from an AI assistant.

Operational in 30 seconds

Deploy

Add Maintenant to your docker-compose.yml. Mount the Docker socket read-only. That's it.

docker compose up -d

Auto-discovery

Maintenant detects your containers, their health checks, Compose projects. On Kubernetes, it detects the runtime automatically.

5 containers discovered

Monitor and alert

Add Docker labels for HTTP/TCP endpoints. Configure your Discord or Slack webhooks. Your stack is under control.

0 incidents · 99.98% uptime

Copy. Paste. Monitor.

# docker-compose.yml
services:
  maintenant:
    image: ghcr.io/kolapsis/maintenant:latest
    ports:
      - "8080:8080"
    volumes:
      - /var/run/docker.sock:/var/run/docker.sock:ro
      - maintenant-data:/data
    environment:
      MAINTENANT_DB: "/data/maintenant.db"
    restart: unless-stopped
volumes:
  maintenant-data:

Compatible with Docker Engine 20.10+ and Kubernetes · Linux amd64/arm64 · macOS · WSL2

One tool instead of five

MaintenantUptime KumaGrafana StackHealthchecksPortainer
Container auto-discovery
HTTP/TCP Checks
Heartbeats / Cron
SSL Monitoring
CPU/RAM/Network metricsLimited
Update detection
Public status page
Native Kubernetes
Single binary, zero depsNode.js3+ containersPython

Simple. Transparent.

100% on-premise. Your data stays with you. Always.

Community
Free forever
Open source · AGPL-3.0
  • Unlimited container auto-discovery
  • HTTP/TCP endpoint monitoring
  • Heartbeats / cron (10 max)
  • SSL/TLS certificates
  • CPU, RAM, network metrics
  • Update Intelligence (digest scan)
  • Webhook + Discord
  • Public status page
  • REST API + SSE + MCP
  • PWA
Install for free

Your stack deserves
to be monitored.

One container. Zero config. Install Maintenant and take back control.