Monitor a personal VPS or home server without sending data to any third-party cloud service
Deploy a lightweight monitoring dashboard on a small server using a one-line install script
Run the Komari agent on multiple machines and view all their metrics in one dashboard
Build a custom agent client that reports metrics from a non-standard device to the Komari server
One-click install script works on Linux systems with systemd, Docker option also available.
Komari is a lightweight, self-hosted server monitoring tool. You run it on your own server, and it gives you a web dashboard where you can see how that server is performing, including metrics collected by a small agent program. Because you host it yourself, your data stays under your control and does not go through any third-party service. The tool is built to be low on resources, so it works well even on small or modest servers. The web interface is described as an intuitive monitoring dashboard, though the README does not detail which specific metrics are displayed beyond linking to external documentation. Getting started has three main paths. The quickest is a one-click install script that works on Linux systems using systemd, the init system used by Ubuntu and Debian among others. The second option is Docker, where you run a container image, point it at a local data folder, and access the dashboard at a specific port in your browser. The third is downloading a prebuilt binary file and running it directly. A manual build option also exists for developers, which requires Go and Node.js and involves building both a frontend and a backend separately. The project supports themes and has a development guide for creating custom agent clients that report data to the server. Documentation is hosted on an external site. A Telegram group is available for community support. Komari is written in Go and the source is open. Contributions via issues and pull requests are welcome.
← komari-monitor on gitmyhub — every repo by this author, as a profile.
Verify against the repo before relying on details.