Ask your homelab what's wrong

Type "why is pve-storage slow?" and get a plain-English answer. 8MB Go agent, open source, 5-minute setup.

Install Free View Dashboard Demo
$ curl -fsSL https://labwatch.zazastation.duckdns.org/install.sh | sudo bash copy

Or download the binary directly from amd64 / arm64. Open source — AGPL-3.0 server, MIT agent. Source on GitHub. (review the script first)

See your lab at a glance

labwatch fleet dashboard labwatch node detail view

Fleet overview (left) and node detail (right) — real data from a 5-node Proxmox cluster.

Three steps. Five minutes.

From zero to full homelab intelligence in one terminal session.

1

Install the agent

One command installs a lightweight agent on each node. No inbound ports to open, under 50MB RAM at runtime.

2

Metrics flow automatically

CPU, memory, disk, network, Docker containers, services. Outbound only — your firewall stays closed.

3

Ask questions, get answers

Type questions in plain English. Get smart alerts, metric diagnostics, and daily intelligence digests with per-node grades.

Not just graphs. Intelligence.

labwatch understands your homelab, not just its metrics.

Natural Language Queries

Type "why is pve-storage slow?" and get a diagnostic answer. Not a graph — an explanation with metric context.

NO LLM — pure regex + rules

Smart Alerts

Deduplication, auto-resolution, and threshold monitoring. Catches slow disk fills and resource pressure before they become outages.

Fleet Intelligence

Daily plain-English digest of your entire lab. Per-node letter grades, trends, anomalies, and actionable recommendations.

Docker Monitoring

Container health, restart loop detection, CPU and memory per container. Sees inside your stacks, not just the host.

Works With Your Stack

Runs alongside Grafana, Uptime Kuma, Prometheus. labwatch adds the intelligence layer they don't have.

Open Source, Your Control

MIT-licensed Go agent you can audit. Outbound-only connections. Self-host the server or use our hosted version. Your data stays yours.

Compared to Beszel, Uptime Kuma & Netdata

Honest feature matrix. Everyone's open source. Each tool is good at something different.

Beszel wins on pure minimalism. Netdata wins on feature breadth. Kuma wins on uptime focus. labwatch wins when you want plain-English answers from your own metrics — no LLM, no cloud.

Feature labwatch Beszel Uptime Kuma Netdata
Natural-language queries ("why is pve slow?") ✓ ✕ ✕ cloud only
Multi-host agents ✓ ✓ indirect ✓
Per-user dashboard layout (drag, pin, reorder) ✓ ✕ basic basic
24h heartbeat bar ✓ ✕ ✓ ✕
Anomaly dots on sparklines ✓ ✕ ✕ ✓
One-command install ✓ ✓ ✕ ✓
Click-to-test notifications ✓ ✕ ✓ ✕
Agent RAM (typical steady-state) ~20 MB ~15 MB (server only) ~150 MB
Install complexity 1 command 1 command docker 1 command
Docker container stats ✓ ✓ ✕ ✓
GPU + temperature ✓ ✓ ✕ ✓
S.M.A.R.T. disk health (temp, reallocated sectors, pass/fail) ✓ ✓ ✕ ✓
ZFS pool health (capacity, scrub status, errors) ✓ partial ✕ ✓
Disk I/O metrics (read/write throughput, IOPS) ✓ ✓ ✕ ✓
Centralized log collection (journald + Docker, searchable) ✓ ✕ ✕ Loki needed
Notification channels (integrations) ntfy, Discord, Slack, Telegram + Apprise Shoutrrr (~20) 90+ ~15
Prometheus / Grafana export (/metrics endpoint) ✓ ✓ ✕ ✓
HTTP / TCP uptime checks (external probe) local ✕ ✓ ✕
Metric collectors / plugins built-in built-in N/A 800+
Open source ✓ ✓ ✓ ✓

See it in action

Real queries, real answers. No graphs to interpret.
No LLM, no API keys, no data leaves your server. 100% rule-based engine.

labwatch query
> why is pve-storage slow? _
pve-storage shows I/O pressure. Load average (38.1) is high relative to CPU count (2) but CPU usage is low (4.2%). This pattern indicates disk or network I/O bottleneck — processes waiting on I/O rather than computing. Memory at 12.1%, Disk at 55.3% — no other resource pressure.

Simple, honest pricing

Start free. Scale when you need to.

Free
$0
Everything, capped at 3 nodes. No credit card.
  • + 3 nodes
  • + 30-day retention
  • + Smart alerts + notifications
  • + Natural language queries
  • + Intelligence digest
Get Started
Pro
$9/mo
For serious labs — unlimited nodes and priority.
  • + Unlimited nodes
  • + 1 year retention
  • + Everything in Free
  • + Priority support
  • + We host it for you
Get Started
Business
Custom
For teams running 50+ servers who need SLA-backed support and custom integrations.
  • + Unlimited nodes
  • + Custom integrations
  • + SLA & priority support
  • + Self-hosted option
  • + Dedicated onboarding
Book a call