The monitoring stackyou own outright.
Uptime, pagespeed, hardware and incident alerts in one self-hosted dashboard. No per-monitor pricing. And you find out before your users do.
Demo credentials: demouser@demo.com / Demouser1!
Monitoring shouldn't costa fortune.
Enterprise tools meter every host. Free tools hit a wall fast. Checkmate sits in the middle: self-hosted, fully featured and not trying to upsell you at 2am.
Per-host pricing adds up
Datadog and PagerDuty charge per host, per seat and per feature. Checkmate has one tier: all of it, free.
Your metrics, their servers
SaaS monitoring ships telemetry off your network. Self-host Checkmate and nothing leaves your perimeter. There's no third party to trust, because there's no third party.
Stacks that take a week
Prometheus, Grafana, Alertmanager, exporters. Checkmate gets you to a first alert in about 5 minutes.
10 monitor types,one dashboard.
An endpoint can return 200 while the cert expires, the disk fills and a container crash-loops. A green dot can't see any of that. These can.
Everything you need,under one roof.
Uptime, pagespeed, servers, containers and alerts behind a single login. Outages love the gaps between tools, so we didn't leave any.
Global uptime monitoring
Monitor HTTP, ping and TCP endpoints from 6 continents with GlobalPing. Know when things go down, anywhere in the world.
Server monitoring
Track CPU, memory, disk and network usage across all your servers.
Docker monitoring
Know the moment a container stops, with health and resource views on the way.
Multi-channel alerts
Get notified on 12 channels, from email to PagerDuty. You hear about it before your users do.
Public status pages
Beautiful, customizable status pages to keep your users informed during incidents.
Game server monitoring
Monitor Minecraft, CS2, Valheim and more than 100 game types.
Response time tracking
Track response times and latency with historical charts and performance insights.
Team collaboration
Invite your team with role-based access control. Everyone stays in the loop.
The dashboard,as it ships.
These are the four views you'll spend the most time in. Each one loads fast, works from the keyboard and shows you what changed without making you dig for it.
Know the moment something breaks.
Per-endpoint uptime, incident counts and response time history. Chart views span a day, week, or month so patterns show up before they become outages.
- Real-time uptime percentage and incident tracking
- Response time charts with day, week and month views
- SSL certificate expiry monitoring
Lighthouse scores, tracked over time.
Performance, accessibility, SEO, best practices, plus Core Web Vitals. Every run recorded so regressions don't ship silently.
- Lighthouse performance, accessibility and SEO scores
- Core Web Vitals: LCP, FCP, CLS and more
- Score history to track improvements over time
Server health, down to the core.
CPU, memory, disk and network across every box, reported by the Capture agent. Live gauges for right-now, history for capacity planning.
- CPU usage, temperature and core frequency
- Memory and disk usage with visual gauges
- Historical charts for capacity planning
A status page users trust.
Public dashboard with per-service uptime bars and your own branding. One shareable URL for customers, on-call and stakeholders.
- Public-facing status dashboard with custom branding
- Per-service uptime bars with historical data
- Shareable link for customers and stakeholders
Every outage, on the record.
Checkmate opens an incident on the first failed check, tracks how long it lasts and keeps the timeline after recovery. The history is your SLA math and your postmortem notes in one place.
- Incidents created automatically from failed checks
- Timelines with start, duration and resolution
- Filter by monitor, date range or status
Planned downtime, without the noise.
Schedule one-off or recurring windows and Checkmate holds the alerts while you work. Checks keep running and get recorded; nobody gets paged for the reboot you planned.
- One-off and recurring schedules
- Alerts suppressed, checks still recorded
- Windows scoped to the monitors you pick
Plugs into the toolsyou already run.
Who runsCheckmate?
Homelabs
A Pi and a compose file watching the NAS, the reverse proxy and 40 containers.
SaaS founders
One founder, one server and a status page that answers "is it down?" before support has to.
Agencies
Every client site in one dashboard, with a branded status page for each.
Sysadmins and DevOps
TCP, DNS, ping and hardware metrics without standing up a Prometheus stack.
E-commerce
Checkout uptime during the sale, certificate expiry visible long before it matters.
Open-source maintainers
Free monitoring for project infra, run on the project's own box.
Your servers,your rules.
Docker Compose
A single compose file. Up and running on your box in under 5 minutes.
Bare metal or k8s
Drop the containers on a VM, a Pi, or a Helm chart. No artificial monitor limits.
AGPL-3.0
Source is yours to read, fork and audit. Data never leaves your perimeter.
Built in the open,for everyone.
AGPL-3.0, no vendor lock-in, no paywalled features. Read the code, file an issue, or send a PR. 130-plus contributors already have. It has run in homelabs and prod racks since 2024.
Shipped byreal humans.
Frequently askedquestions.
A Linux box (or any Docker host) with about 1 GB of RAM and a few GB of disk. Checkmate ships as a small set of containers: pull the compose file, run it, done.
AGPL-3.0. You can run it for free, fork it and modify it. If you expose a modified version as a network service, you need to share your changes under the same license.
No. Everything (metrics, logs, incidents, user accounts) lives in the database you point Checkmate at. No phone-home, no telemetry pipeline.
A ping tells you the server answered. It can't tell you the cert is valid next week, DNS still points where it should, the disk isn't filling or a container isn't restart-looping. That's why there are 10 monitor types instead of one.
Commercial tools meter by host or monitor. Checkmate doesn't. You give up the managed cloud and some polish; you get unlimited monitors, your own data and a smaller bill.
There's no one-click import, but most existing monitors map cleanly: HTTP, ping, TCP, Docker, status pages. For bulk setup the API accepts JSON, so a short script usually does it.
Discord is the fastest path; maintainers and contributors hang out there. GitHub issues work for bugs and feature requests. There is no paid support tier.
Stop worryingabout downtime.
A single compose file, a few minutes and the next incident reaches you first, not your users. Free forever, self-hosted.