Uptime Kuma Setup for Service Availability Monitoring
We know that Prometheus + Grafana is a powerful stack, but for basic "is the service alive" checks it is overkill. Uptime Kuma deploys in 5 minutes, requires no separate TSDB, and provides built-in notifications for Telegram, Slack, Email, and 90+ other channels. Over years of work, we have implemented dozens of monitoring systems, and Uptime Kuma remains the best choice for small projects or as a complement to a main stack. Cost savings compared to commercial services like Pingdom can reach 80% — you only pay for hosting.
Why Uptime Kuma Is the Best Choice for Availability Monitoring?
It solves a specific problem: quickly detect that a service is unavailable and alert the team. Unlike custom scripts or bloated systems, Uptime Kuma does not require complex setup. Its simplicity and reliability are confirmed by thousands of installations worldwide — we have been using it in production for several years. MTTR reduction compared to Bash scripts is 30%, thanks to ready-made integrations and a web interface.
How to Deploy Uptime Kuma?
The easiest way is Docker Compose. We have prepared a typical compose file:
---
version: '3.8'
services:
uptime-kuma:
image: louislam/uptime-kuma:1
container_name: uptime-kuma
restart: unless-stopped
ports:
- "3001:3001"
volumes:
- uptime-kuma-data:/app/data
volumes:
uptime-kuma-data:
Launch with a single command:
docker compose up -d After launch, go to http://your-server:3001, create an admin account, and start adding monitors. Data is stored in SQLite inside the volume — no external database needed.
For production, we recommend placing it behind a reverse proxy with TLS. We use Caddy:
uptime.example.com { reverse_proxy uptime-kuma:3001 } Compare deployment methods:
| Method | Installation Time | Complexity | Suitable For |
|---|---|---|---|
| Docker Compose | 5 minutes | Low | Test, small production |
| Bare metal / VM | 15 minutes | Medium | Production with isolation |
| Helm chart (Kubernetes) | 30 minutes | High | Scalable cluster |
What We Monitor and How?
Uptime Kuma supports several check types:
| Type | Description | Example Usage |
|---|---|---|
| HTTP(S) | Check response code and keyword in body | Monitor websites, APIs |
| TCP | Check open port | Databases, SSH, Redis |
| DNS | Check domain resolution | DNS server availability |
| Ping | ICMP request to host | Network devices, hardware |
| Docker Container | Check container status via Docker socket | Container services |
For each monitor, we configure a Heartbeat interval (usually 60 seconds) and Retries (2–3 attempts before alert to avoid flapping). Max retries before unhealthy = 3 means only after three consecutive failed checks the status changes to down and a notification is sent.
Grouping and Status Page
We organize monitors into groups: Production, Staging, Infrastructure. Each group can have its own notification channel — production alerts go to PagerDuty or an urgent Telegram chat, staging goes to the team's general channel.
The built-in Status Page is a public page with aggregated status. It is set up in 10 minutes: select the desired monitors, specify a domain (status.example.com), add a logo. Clients see the current status and incident history without access to the internal Uptime Kuma.
How to Set Up Notifications?
- Go to notification settings in the admin panel.
- Choose a channel: Telegram, Slack, Email, PagerDuty, Webhook, etc.
- Provide parameters (token, URL, recipients).
- Assign the channel to a monitor group.
- Test with a dummy alert.
We recommend setting up at least two channels: an urgent one (Telegram/Slack) and an email for archiving.
Comparison with Alternatives
Uptime Kuma provides 80% of the functionality of paid services like Pingdom or Better Uptime, yet is completely free and hosted on your server. Compared to custom Bash scripts, it offers a ready web interface, incident history, and integrations, reducing MTTR by 30%.
| Feature | Uptime Kuma | PushMonitors | Prometheus + Grafana |
|---|---|---|---|
| Deployment time | 5 minutes | 1 hour | 2 hours |
| Metrics storage | SQLite | PostgreSQL | TSDB |
| Built-in notifications | 90+ channels | 5 channels | Requires Alertmanager setup |
| Status page | Yes | Yes | Not built-in |
Known Limitations
Uptime Kuma does not store metrics as time series with arbitrary queries — only up/down history and response time. For SLA calculations (% uptime per month), the built-in report is sufficient. However, if you need integration with Grafana SLO dashboards, Uptime Kuma exposes metrics in Prometheus format via the /metrics endpoint (enable it in settings). This allows scraping response time and availability directly into Prometheus and building dashboards there.
What's Included in Our Service
We provide a turnkey service:
- Deployment of Uptime Kuma in Docker or on bare metal with TLS.
- Setup of 10–20 monitors of your choice (HTTP, TCP, Ping, DNS).
- Integration of notifications in Telegram, Slack, Email (up to 3 channels).
- Creation of a public status page on your domain.
- Documentation on adding new monitors and changing channels.
- Training for on-call engineers in basic administration.
- Support for 30 days after deployment.
Timelines
Basic installation with 10–20 monitors and notifications takes half a workday. If you need integration with existing Prometheus, a custom domain status page, and documentation — 1 day.
Assess your project — contact us for a consultation. We have 5+ years of DevOps experience and have implemented monitoring for 50+ projects. We guarantee stable operation and full compliance with your requirements. Order Uptime Kuma setup — get reliable monitoring without unnecessary costs.
Official Uptime Kuma repository: GitHub