Headless Servers

The service daemon coolercontrold exposes a web UI and does not require a desktop environment. By default, the daemon binds to the local loopback interface (localhost) on port 11987. The web UI is served directly from the daemon at http://localhost:11987/ and if exposed on an external interface, the web UI is served at https://<daemon-ip>:<daemon-port>/ using a self-signed certificate (or a certificate of your choice if you configure SSL/TLS). Additionally, a gRPC API is exposed on port 11988 and includes a standard health check endpoint. The protobuf spec is available at /proto/coolercontrol.

Accessing the Web UI on Headless Servers

Once the daemon's IP address and port are configured so that you can reach it remotely, you can access the CoolerControl web UI from another machine. This is the recommended approach for controlling your headless server's cooling devices. You can make many changes to the daemon's config.toml file, but due to the advanced nature of CoolerControl's control options and various entity relationships, it is highly recommended to use the web UI for cooling control.

TIP

For systems with an NVIDIA GPU, ensure the proprietary driver is installed. CoolerControl uses NVML and the CLI tools nvidia-settings/nvidia-smi when available. On fully headless setups (no Xorg/Wayland), fan control via the CLI tools may not be available.

TIP

Running CoolerControl in Docker? Containers are inherently headless and the Docker image is pre-configured to bind to all interfaces with TLS disabled. See the Docker installation guide.

Headless Servers ​

Accessing the Web UI on Headless Servers ​

See Also ​

Headless Servers

Accessing the Web UI on Headless Servers

See Also