Skip to content

Configuration tutorial of MassGrid Docker Edge Node

Jump to bottom
Wany edited this page Jul 16, 2019 · 3 revisions

MassGrid边缘节点配置教程v1.3.3.1

准备 1台带有NVIDIA显卡机器

最低配置如下：

Cpu: Intel Celeron N3350

Memory: ddr3 4G

Gpu: P106-100

ssd/hdd: 60G

1、安装ubuntu16.04

ubuntu16.04 下载链接

制作系统可使用软碟通做u盘启动盘，在此不做赘述。

需要注意的是：配置分区时候需要预留一个空分区（50G以上）以备docker使用。

如下图:

系统分区(sda3) 40G

docker分区(sda4) 200G

分区配置

2、安装NVIDIA驱动和CUDA

1)下载CUDA10.0文件

下载地址

将下载后的文件 放到 $HOME/nvidia 目录下

以下用户以mgd ，驱动文件为cuda_10.0.130_410.48_linux.run为例

2) 禁用 nouveau
echo "blacklist nouveau" | sudo tee -a /etc/modprobe.d/blacklist.conf
sudo update-initramfs -u
#最后sudo reboot 重启机器
sudo reboot
3) 按ctrl + alt +F1 进入tty1
sudo service lightdm stop #停止桌面
sudo /home/mgd/nvidia/cuda_10.0.130_410.48_linux.run --no-opengl-libs --driver --silent
#安装驱动和cuda
nvidia-smi

image-20190711112645454

可以看到驱动已经安装成功

sudo service lightdm start #恢复桌面

3、安装docker

使用ubuntu16.04的一键脚本安装docker

wget https://raw.githubusercontent.com/wanyvic/DockerImageBase/master/docker_install.sh
chmod +x docker_install.sh
sudo ./docker_install.sh

##执行成功则可以看到具体的docker信息
Client:
 Version:           18.06.1-ce
 API version:       1.38
 Go version:        go1.10.3
 Git commit:        e68fc7a
 Built:             Tue Aug 21 17:24:56 2018
 OS/Arch:           linux/amd64
 Experimental:      false

Server:
 Engine:
  Version:          18.06.1-ce
  API version:      1.38 (minimum version 1.12)
  Go version:       go1.10.3
  Git commit:       e68fc7a
  Built:            Tue Aug 21 17:23:21 2018
  OS/Arch:          linux/amd64
  Experimental:     false

4、配置文件接入主节点

1) 下载一键配置脚本
wget https://raw.githubusercontent.com/wanyvic/DockerImageBase/master/dockerConfig.py
wget https://raw.githubusercontent.com/wanyvic/DockerImageBase/master/setDockerToken.sh
chmod +x setDockerToken.sh
2)配置接入主节点 token

以接入主节点ip 49.234.37.251为例

MassGrid 控制台输出

masternodelist full 49.234.37.251

{
  "95558da5ca304ddb43743cdea4c2ea9957f32de6f5f9ac5ba082d5fd71268e60-0": "           ENABLED 70031 M9jqzZUg3NRNCbSWZxhJyYny4sasYVHH1f 1562815853  1291335 1562794970 181503 49.234.37.251:9443 8 1 6 10061 302383360 SWMTKN-1-0uau6cvrqjv90wx0ka4h5g04bqwe0t0lmy25secg8i81rg15uj-dcoko0h4rlfdnkw9kd5z7vsn4 49.234.37.251:2377"
}

获取SWMTKN开头的 token: SWMTKN-1-0uau6cvrqjv90wx0ka4h5g04bqwe0t0lmy25secg8i81rg15uj-dcoko0h4rlfdnkw9kd5z7vsn4 49.234.37.251:2377

3)设置闲时挖矿
#demo1
sudo python dockerConfig.py -all -pf -miner "mgd" -addr "MTzzXdhT3NDyfFLUL42bVYeewYpt8JSqAm" -revenue "MTzzXdhT3NDyfFLUL42bVYeewYpt8JSqAm" -dev "sda4" -join "SWMTKN-1-0uau6cvrqjv90wx0ka4h5g04bqwe0t0lmy25secg8i81rg15uj-dcoko0h4rlfdnkw9kd5z7vsn4 49.234.37.251:2377"
#demo2
sudo python dockerConfig.py -all -pf -miner "eth" -addr "0xff6514c44f4D3AFeEFDFca7913D33Ec3a87747f3" -revenue "MTzzXdhT3NDyfFLUL42bVYeewYpt8JSqAm" -dev "sda4" -join "SWMTKN-1-0uau6cvrqjv90wx0ka4h5g04bqwe0t0lmy25secg8i81rg15uj-dcoko0h4rlfdnkw9kd5z7vsn4 49.234.37.251:2377"

-miner 当前支持 eth、mgd

-addr 闲时挖矿地址

-revenue mgd收款地址

-dev docker分区

-join 主节点token

sudo python dockerConfig.py -load 

#查看机器配置信息
{
    "default-runtime": "nvidia", 
    "labels": [
        "localip=192.168.2.176", 
        "address=mjxR2WQragrPfvhwCLyyUj5FHCb1ULP8DL", 
        "cpucount=4", 
        "cpuname=intel_i3", 
        "memcount=16", 
        "memname=ddr", 
        "gpucount=1", 
        "gpuname=NVIDIA_P104_100_4G", 
        "miner_address=MTzzXdhT3NDyfFLUL42bVYeewYpt8JSqAm", 
        "miner_pool=mgd.vvpool.com:5630", 
        "miner_worker=mgd176", 
        "miner_type=MGD"
    ], 
    "node-generic-resources": [
        "NVIDIA_P104_100_4G=GPU-92205b0"
    ], 
    "runtimes": {
        "nvidia": {
            "path": "nvidia-container-runtime", 
            "runtimeArgs": []
        }
    }, 
    "storage-driver": "devicemapper", 
    "storage-opts": [
        "dm.directlvm_device=/dev/sda4", 
        "dm.thinp_percent=95", 
        "dm.thinp_metapercent=2", 
        "dm.directlvm_device_force=true", 
        "dm.basesize=40G"
    ]
}
Clone this wiki locally