Skip to content

Installation

Jump to bottom
Max edited this page Jun 6, 2024 · 45 revisions

Windows

You will need to install and update "Microsoft Visual C++ Redistributable for Visual Studio 2022" to the latest.

Scroll down to "Other tools and redistributables:" https://visualstudio.microsoft.com/downloads/

Windows executables: https://github.com/madMAx43v3r/mmx-node/releases

Linux

Dependencies

Ubuntu Linux:

sudo apt update
sudo apt install git cmake build-essential automake libtool pkg-config curl libminiupnpc-dev libjemalloc-dev libzstd-dev zlib1g-dev ocl-icd-opencl-dev clinfo screen

Arch Linux:

sudo pacman -Syu
sudo pacman -S base-devel git cmake curl miniupnpc jemalloc zstd zlib opencl-headers ocl-icd clinfo screen

Fedora Linux:

yum install kernel-devel git cmake automake libtool curl gcc gcc-c++ miniupnpc-devel jemalloc-devel ocl-icd-devel zlib-ng-devel zstd clinfo screen

Note: OpenCL packages are optional, ie. ocl-icd-opencl-dev, etc...

OpenCL provides faster and more efficient VDF verification using an integrated or dedicated GPU.

A standard iGPU found in Intel 1165G7 with 96EU iGPU on a low power laptop can verify VDF in about 4 sec at 48 MH/s timelord speed. For more info, please see: What kind of GPU do I need for VDF verifications?

If you dont have a fast modern (>4 GHz) CPU, it is required to have a GPU with OpenCL support.

Make sure to be in the video and or render group (depends on distribution) to be able to access a GPU:

sudo adduser $USER video
sudo adduser $USER render

Building

git clone https://github.com/madMAx43v3r/mmx-node.git
cd mmx-node
./update.sh

To update to latest version:

./update.sh

To switch to latest testnet:

rm NETWORK
./update.sh

Rebuilding

If you suspect some files might not build properly or if you want to recompile, stop the node then run:

./clean_all.sh
./update.sh

This is needed when updating system packages for example.

Windows via WSL

To setup Ubuntu 20.04 in WSL on Windows you can follow the tutorial over here:
https://docs.microsoft.com/en-us/learn/modules/get-started-with-windows-subsystem-for-linux/

Make sure to install Ubuntu in step 2: https://www.microsoft.com/store/p/ubuntu/9nblggh4msv6

Then type "Ubuntu" in the start menu and start it, you will be asked to setup a user and password. After that you can follow the normal instructions for Ubuntu 20.04.

To get OpenCL working in WSL: https://devblogs.microsoft.com/commandline/oneapi-l0-openvino-and-opencl-coming-to-the-windows-subsystem-for-linux-for-intel-gpus/

Docker

Docker images are available via ghcr.io/madmax43v3r/mmx-node in a few flavours:

  • edge: Latest commit cpu only
  • edge-amd: Latest commit with amd gpu support
  • edge-nvidia: Latest commit with nvidia gpu support

Additionally, each semver tag produces tagged images using the same three flavours from above: <major>.<minor>.<patch>, <major>.<minor> and <major> each with their respective suffix (eg. 0.9.5-amd). latest is set for the latest cpu only semver image.

Each image provides a volume for /data which you can override with your own volume or a mapped path to customize the storage location of the node data.

CPU only

A docker-compose.yml for the cpu only node can look like this:

version: '3'
services:
  node:
    image: ghcr.io/madmax43v3r/mmx-node:edge
    restart: unless-stopped
    volumes:
      - /some/path/to/mmx/node/data:/data

AMD GPU

For amd gpu support please see the following docker-compose.yml:

version: '3'
services:
  node:
    image: ghcr.io/madmax43v3r/mmx-node:edge-amd
    restart: unless-stopped
    group_add:
      - video
      - render
    devices:
      - /dev/dri:/dev/dri
      - /dev/kfd:/dev/kfd
    volumes:
      - /some/path/to/mmx/node/data:/data

Note: - render in group_add might need to be removed, depending on your system.

NVIDIA GPU

For nvidia gpu support please see the following docker-compose.yml:

version: '3'
services:
  node:
    image: ghcr.io/madmax43v3r/mmx-node:edge-nvidia
    restart: unless-stopped
    runtime: nvidia
    volumes:
      - /some/path/to/mmx/node/data:/data

Note: for nvidia you also need the NVIDIA Container Toolkit installed on the host, for more info please see: https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/install-guide.html#docker

Remote harvester

Running a remote harvester is done by overwriting the CMD of the Dockerfile, for example like this:

version: '3'
services:
  harvester:
    image: ghcr.io/madmax43v3r/mmx-node:edge
    restart: unless-stopped
    volumes:
      - /some/path/to/mmx/node/data:/data
      - /some/path/to/disks:/disks
    command: './run_harvester.sh -n <some ip or hostname here>:11333'

Custom storage path

To change the storage path for everything you can set environment variable MMX_HOME to /your/path/ (trailing slash required). By default the current directory is used, ie. mmx-node.

Clone this wiki locally