Ubuntu AI-Docker环境搭建

Ubuntu AI-Docker环境搭建

一、解决步骤

1. 更改DNS

   $ sudo su -
   $ mv /home/user/actuate/* ./
   $ systemctl disable --now systemd-resolved
   $ rm /etc/resolv.conf
   $ touch /etc/resolv.conf
   $ echo "nameserver 223.5.5.5" >> /etc/resolv.conf

2. 更改镜像源

   $ vim/etc/apt/sources.list
   deb http://mirrors.aliyun.com/ubuntu/ bionic main restricted universe multiverse
   deb-src http://mirrors.aliyun.com/ubuntu/ bionic main restricted universe multiverse
   
   deb http://mirrors.aliyun.com/ubuntu/ bionic-security main restricted universe multiverse
   deb-src http://mirrors.aliyun.com/ubuntu/ bionic-security main restricted universe multiverse
   
   deb http://mirrors.aliyun.com/ubuntu/ bionic-updates main restricted universe multiverse
   deb-src http://mirrors.aliyun.com/ubuntu/ bionic-updates main restricted universe multiverse
   
   # deb http://mirrors.aliyun.com/ubuntu/ bionic-proposed main restricted universe multiverse
   
   # deb-src http://mirrors.aliyun.com/ubuntu/ bionic-proposed main restricted universe multiverse
   
   deb http://mirrors.aliyun.com/ubuntu/ bionic-backports main restricted universe multiverse
   deb-src http://mirrors.aliyun.com/ubuntu/ bionic-backports main restricted universe multiverse

3. 安装依赖

   $ apt update -y
   $ apt install gcc make -y
   $ apt install m4 graphviz libnl-route-3-dev swig debhelper tk tcl bison chrpath autoconf libltdl-dev quilt gfortran dkms pkg-config autotools-dev dpatch libgfortran4 flex automake libnl-3-dev libnl-route-3-200 python3-distutils python -y

4. 禁用本地驱动

   $ touch /lib/modprobe.d/dist-blacklist.conf
   $ vim /lib/modprobe.d/dist-blacklist.conf
   blacklist nouveau
   options nouveau modeset=0
   $ rmmod nouveau

5. 安装docker 和nvidia-docker2 及驱动

   $ apt -y install docker.io
   $ systemctl stop docker
   $ distribution=$(. /etc/os-release;echo $ID$VERSION_ID) && curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | sudo apt-key add -&& curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list | sudo tee /etc/apt/sources.list.d/nvidia-docker.list
   $ curl -s -L https://nvidia.github.io/nvidia-container-runtime/experimental/$distribution/nvidia-container-runtime.list | sudo tee /etc/apt/sources.list.d/nvidia-container-runtime.list
   $ apt update
   $ apt install -y nvidia-docker2
   $ echo  '{
   
         "default-runtime": "nvidia",
   
         "exec-opts": ["native.cgroupdriver=systemd"],
   
         "runtimes": {
   
           "nvidia": {
   
             "path": "nvidia-container-runtime",
   
             "runtimeArgs": []
   
           }
   
         }
   
       }'  >  /etc/docker/daemon.json
   
   fi
   #安装FW
   $ dpkg -i nvidia-fabricmanager-470_470.103.01-1_amd64.deb
   $ dpkg -i nvidia-fabricmanager-dev-470_470.103.01-1_amd64.deb
   #重启服务器
   $ reboot
   #安装GPU显卡驱动
   $ ./NVIDIA-Linux-x86_64-470.103.01.run --no-opengl-files --ui=none --no-questions --accept-license
   $ systemctl restart nvidia-fabricmanager.service
   $ systemctl status nvidia-fabricmanager.service
   #nvidia-smi -mig 0 解决nvlink不显示问题

6. 测试命令

   $ docker run hub.kce.ksyun.com/minimax-pub/cuda-sample:vectoradd-cuda10.2 

二、Ubuntu LVM 扩展存储（简易流程）

1. 查看可用VG大小

   sudo vgdisplay
   # Alloc PE / SIZE 行第二个数字为已分配大小200 GiB
   # Free PE / SIZE 行第二个数字为未分配大小246.09 GiB
   # 即除去系统占用，当前可用总大小为446 GiB

2. 查看LV名字

   sudo lvdisplay
   # LV PATH 行即是名字，后面改大小会使用

3. 修改LV大小

   sudo lvextend -L 446G /dev/ubuntu-vg/ubuntu-lv
   # 使用lvextend工具 -L 指定修改后的大小以及修改的LV名称

4. 重设LV分区大小

   sudo resize2fs /dev/ubuntu-vg/ubuntu-lv
   # 使用resize2fs工具重设LV分区

5. 完成

三、网络配置

1. Ubuntu 18.04

   # /etc/netplay/0-*yaml
   # This is the network config written by 'subiquity'
   network:
   ethernets:
    ens120f0: {}
    ens120f1: {}
    ib0:
      addresses: [10.103.112.26/16]
   bonds:
    bond4:
      addresses:
      - 10.30.116.31/24
      gateway4: 10.30.116.254
      nameservers:
        addresses: [223.5.5.5]
      interfaces:
        - ens120f0
        - ens120f1
      parameters:
        mode: 802.3ad
        mii-monitor-interval:
        lacp-rate: fast
        transmit-hash-policy: layer3+4
   version: 2