Maintenant — Simpler Alternative to Prometheus + Grafana for Docker Monitoring

The Complexity Problem

Prometheus + Grafana is the gold standard monitoring stack. It powers some of the largest infrastructures in the world. But for a typical Docker Compose stack of 10-50 containers on a VPS, it is massively overkill.

To monitor your containers with the Grafana stack, you need at minimum:

• Prometheus — metrics collection and storage
• cAdvisor — exposes container metrics to Prometheus
• Grafana — visualization and dashboards
• node_exporter — host system metrics (recommended)
• Alertmanager — alert routing (if you want notifications)

That is 3 to 5 extra containers just to monitor the containers you already have. Each one with its own configuration, its own updates, its own failure modes. The irony: your monitoring stack can go down independently of your applications.

And then there is PromQL. Writing an alert rule for “my Redis container uses more than 80% of its allocated memory” requires:

container_memory_usage_bytes{name="redis"} /
container_spec_memory_limit_bytes{name="redis"} > 0.8

With Maintenant, you add a Docker label and the default thresholds handle the rest.

Who Should NOT Switch

Maintenant does not replace Prometheus + Grafana if you need:

• Ad-hoc queries — PromQL lets you freely explore and correlate any metric. Maintenant has pre-built views with no custom query language.
• Custom dashboards — Grafana’s dashboard builder is unmatched. Maintenant’s views are opinionated and fixed.
• Long-term metric retention — Prometheus can store years of high-resolution data. Maintenant retains 7 days (Community) or 1 year (Pro).
• Multi-cluster federation — Prometheus federation allows aggregating metrics across clusters. Maintenant is per-host or per-cluster.
• Application-level metrics — if your Go/Java/Python app exposes custom Prometheus metrics, Maintenant does not scrape them.

If you have a dedicated SRE team and hundreds of microservices, stick with Prometheus + Grafana.

Who Should Switch

• Self-hosters who spent more time configuring Prometheus than running their actual services
• Small teams who need “is everything running?” not “what is the 95th percentile of request duration across my service mesh?”
• Developers who want monitoring without learning PromQL, writing Grafana dashboards, or maintaining a 5-container monitoring stack
• Homelab enthusiasts tired of cAdvisor eating 200+ MB of RAM just to expose container metrics

Maintenant vs Prometheus + Grafana — Feature Comparison

The Numbers

What You Get with Maintenant

One container. Zero config files. Everything monitored:

• Container auto-discovery — Docker and Kubernetes, no targets to declare

• HTTP/TCP endpoint monitoring — configured via Docker labels, no Blackbox Exporter needed

• SSL/TLS certificate tracking — automatic detection, no separate exporter
• Heartbeat / cron job monitoring — built-in, no external tool needed (10 free, unlimited with Pro)
• System resources — CPU, RAM, network, disk per container and per host

• Update intelligence — OCI digest scan across all containers

• Unified alerts — webhook, Discord (Community), Slack, Teams, Email (Pro)
• Public status page — real-time, SSE updates (Pro: incident management, maintenance windows)

Migrating from Prometheus + Grafana

No data migration needed. Deploy Maintenant alongside your existing stack, verify it covers your needs, then remove Prometheus, Grafana, cAdvisor, and the rest:

# docker-compose.yml — replace 5 containers with 1
services:
  maintenant:
    image: ghcr.io/kolapsis/maintenant:latest
    ports:
      - "8080:8080"
    volumes:
      - /var/run/docker.sock:/var/run/docker.sock:ro
      - /proc:/host/proc:ro
      - maintenant-data:/data
    environment:
      MAINTENANT_ADDR: "0.0.0.0:8080"
      MAINTENANT_DB: "/data/maintenant.db"
    restart: unless-stopped

volumes:
  maintenant-data:

docker compose up -d

Open http://your-server:8080 — your containers are already there. No prometheus.yml. No Grafana datasources. No PromQL.

FAQ

Can I run Maintenant alongside Prometheus during a transition? Yes. Maintenant is read-only and does not interfere with any other monitoring tool.

I have custom Prometheus metrics in my apps. Can Maintenant scrape them? No. Maintenant monitors the infrastructure layer (containers, endpoints, certificates, resources), not application-level metrics. If you need both, you can keep Prometheus for app metrics and use Maintenant for everything else.

What about Grafana Cloud or Prometheus-as-a-Service? If you are already paying for a managed Prometheus service, Maintenant offers a radically simpler (and cheaper) alternative for infrastructure monitoring. But it does not replace application observability.

Does Maintenant support Kubernetes? Yes. It auto-detects the runtime and uses read-only RBAC to discover workloads.