How to Run a Miner Node

Miner nodes are integral to our network, taking on tasks and earning rewards through inference.

Please note that miner nodes are still in heavy development and subject to substantial changes. The current GPU mining rewards are 300 $CAI per GPU per day.

Stable Diffusion Miner

Minimum Hardware Configuration

Component	Requirement
GPU	NVIDIA L4, 3080
RAM	8-16 GB
CPU	1 core
Storage	Depends on traffic volume

Getting Started

Follow these steps to set up and run your Stable Diffusion Miner:

Clone the Repository

git clone https://github.com/cuckoo-network/stable-diffusion-miner-docker.git

Navigate to the Project Directory
```
cd stable-diffusion-miner-docker
```
Download the Necessary Files
```
make download
```
Start the Miner

Add your private key to the command below and start the miner:
```
ETH_PRIVATE_KEY="" make start
```

Ensure you have the required hardware and follow the setup instructions carefully. Stay tuned for updates as we continue to develop and enhance the miner node functionality.

How to setup for Bare Metal Ubuntu Server?

Bare Metal Ubuntu Server

Install Nvidia Container Toolkit

If you encounter the following error when running make start:

[+] Running 1/2
 ✔ Container webui-docker-relay-node-1  Running                                                                                                                                             0.0s
 ⠹ Container webui-docker-auto-1        Starting                                                                                                                                            0.3s
Error response from daemon: failed to create task for container: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: error during container init: error running hook #0: error running hook: exit status 1, stdout: , stderr: Auto-detected mode as 'legacy'
nvidia-container-cli: initialization error: load library failed: libnvidia-ml.so.1: cannot open shared object file: no such file or directory: unknown
make: *** [Makefile:11: start] Error 1

It means the Nvidia Container Toolkit is not installed. Follow the official instructions to install the toolkit.

Custom Docker Daemon Configuration

To use a custom configuration file for Docker, follow these steps:

Prepare Custom Configuration File Ensure your custom configuration file is located at $HOME/.config/docker/daemon.json.
Modify Docker systemd Service If the daemon.json file contains nvidia but running sudo docker run --rm --runtime=nvidia --gpus all ubuntu nvidia-smi results in docker: Error response from daemon: unknown or invalid runtime name: nvidia., modify the Docker systemd service file:
Create a systemd drop-in directory for the Docker service:
```
sudo mkdir -p /etc/systemd/system/docker.service.d
```

Create or edit the override.conf file in this directory:

sudo nano /etc/systemd/system/docker.service.d/override.conf

Add the following configuration to specify the custom config file path:
```
[Service]
ExecStart=
ExecStart=/usr/bin/dockerd --config-file=/home/your-username/.config/docker/daemon.json
```
Replace your-username with your actual username. Use the full path instead of $HOME.
Apply the Changes Reload the systemd manager configuration and restart Docker:
```
sudo systemctl daemon-reload
sudo systemctl restart docker
```
Verify Configuration Check if Docker is using your custom configuration:
```
sudo docker run --rm --runtime=nvidia --gpus all ubuntu nvidia-smi
```

Troubleshooting: Failed to Initialize NVML

If you encounter Failed to initialize NVML: Unknown Error, follow these steps:

Edit the Nvidia container runtime configuration:
```
sudo vim /etc/nvidia-container-runtime/config.toml
```
Change no-cgroups to false and save the file.
Restart the Docker daemon:
```
sudo systemctl restart docker
```

Test the configuration:

sudo docker run --rm --runtime=nvidia --gpus all ubuntu nvidia-smi

How to Run a Miner Node

Stable Diffusion Miner​

Minimum Hardware Configuration​

Getting Started​