Maintenant — Lightweight Alternative to Netdata for Docker Monitoring

What Netdata Does Well

Netdata is a comprehensive real-time monitoring tool with an impressive feature set. It collects thousands of metrics per second, has beautiful auto-generated dashboards, and provides deep visibility into system performance. The per-second granularity is unmatched.

For deep system performance analysis and debugging, Netdata is incredibly powerful.

The Weight Problem

Netdata’s power comes at a cost:

• RAM usage: 300+ MB at rest, often 500+ MB under load. For a VPS with 2 GB RAM running your actual services, that is a significant portion of your resources consumed by monitoring alone.
• CPU usage: Collecting thousands of metrics per second is not free. Netdata can consume 5-15% of a single CPU core continuously.
• Telemetry: Netdata collects anonymous usage statistics by default. While it can be disabled, many self-hosters are uncomfortable with a monitoring tool that phones home.
• Netdata Cloud: The product increasingly pushes users toward Netdata Cloud, their SaaS platform. Some features require a Cloud account.

For a self-hoster running a homelab or a small production stack, Netdata can consume more resources than some of the services it monitors.

Who Should NOT Switch

Maintenant does not replace Netdata if you need:

• Per-second metric granularity — Netdata collects metrics every second. Maintenant collects at configurable intervals (default 30-60s).
• Deep system internals — Netdata monitors 2,000+ system metrics (kernel, networking stack, disk I/O queues, etc.). Maintenant monitors the essentials: CPU, RAM, disk, network.
• Application-level monitoring — Netdata has collectors for databases, web servers, message queues. Maintenant monitors the infrastructure layer.
• Machine learning anomaly detection — Netdata includes built-in ML for anomaly detection. Maintenant does not.

Who Should Switch

• Homelab users whose monitoring tool uses more RAM than their actual applications
• VPS users on limited resources (1-4 GB RAM) where 300+ MB for monitoring is too much
• Privacy-conscious users who want zero telemetry and no cloud account requirements
• Teams who need HTTP endpoint monitoring, SSL tracking, and cron monitoring — things Netdata does not do well
• Docker-first users who want container-aware monitoring with auto-discovery and Docker labels

Maintenant vs Netdata — Feature Comparison

The Numbers

What You Get with Maintenant

Focused monitoring that covers what matters, at a fraction of the resource cost:

• Container auto-discovery — Docker and Kubernetes, zero config

• System resources — the metrics that matter: CPU, RAM, network, disk

• HTTP/TCP endpoint monitoring — active probing with latency tracking

• SSL certificates, heartbeats, update detection, alerts, status page — all built-in
• Network security insights — open ports, exposed services

Migrating from Netdata

Deploy Maintenant, verify it covers your needs, then remove Netdata and free up 300+ MB of RAM:

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:

FAQ

Will I lose visibility by switching from Netdata? You will lose per-second granularity and deep kernel metrics. You will gain HTTP endpoint monitoring, SSL tracking, cron monitoring, update detection, and a public status page — while freeing up 300+ MB of RAM.

Does Maintenant phone home? Never. Zero telemetry. No cloud account. No anonymous statistics. Your data stays on your server.

Can I run both during a transition? Yes, but be mindful of resource usage — running both temporarily will use even more RAM. We recommend testing on a non-production server first.

Netdata has per-second metrics. How often does Maintenant collect? Maintenant collects container and system metrics at configurable intervals, typically every 30-60 seconds. For most self-hosted setups, this is more than sufficient. Per-second granularity is mainly useful for performance debugging, which is not Maintenant’s focus.