docker2、k8s安装

一：安装驱动 [root@localhost home]# chmod 755 NVIDIA-Linux-x86_64-515.48.07.run cuda_10.2.89_440.33.01_linux.run [root@localhost home]# ./NVIDIA-Linux-x86_64-515.48.07.run [root@localhost home]# nvidia-smi Tue Nov 29 16:51:53 2022 +-----------------------------------------------------------------------------+ | NVIDIA-SMI 515.48.07 Driver Version: 515.48.07 CUDA Version: 11.7 | |-------------------------------+----------------------+----------------------+ | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. | | | | MIG M. | |===============================+======================+======================| | 0 NVIDIA GeForce ... Off | 00000000:01:00.0 Off | N/A | | 44% 45C P0 111W / 350W | 0MiB / 24576MiB | 0% Default | | | | N/A | +-------------------------------+----------------------+----------------------+ | 1 NVIDIA GeForce ... Off | 00000000:02:00.0 Off | N/A | | 45% 49C P0 105W / 350W | 0MiB / 24576MiB | 0% Default | | | | N/A | +-------------------------------+----------------------+----------------------+ +-----------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=============================================================================| | No running processes found | +-----------------------------------------------------------------------------+ 二：安装cuda10.2 [root@localhost home]# ./cuda_10.2.89_440.33.01_linux.run [root@localhost home]# vim ~/.bashrc export PATH=/usr/local/cuda-10.2/bin${PATH:+:${PATH}} export LD_LIBRARY_PATH=/usr/local/cuda-10.2/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}} [root@localhost home]# source ~/.bashrc [root@localhost home]# nvcc -V nvcc: NVIDIA (R) Cuda compiler driver Copyright (c) 2005-2019 NVIDIA Corporation Built on Wed_Oct_23_19:24:38_PDT_2019 Cuda compilation tools, release 10.2, V10.2.89 三：安装cudnn8.3.3 [root@localhost home]# tar -xvf cudnn-linux-x86_64-8.3.3.40_cuda10.2-archive.tar.xz [root@localhost home]# cd cudnn-linux-x86_64-8.3.3.40_cuda10.2-archive/ [root@localhost cudnn-linux-x86_64-8.3.3.40_cuda10.2-archive]# cp include/cudnn*.h /usr/local/cuda/include [root@localhost cudnn-linux-x86_64-8.3.3.40_cuda10.2-archive]# cp -P lib/libcudnn* /usr/local/cuda/lib64 [root@localhost cudnn-linux-x86_64-8.3.3.40_cuda10.2-archive]# chmod a+r /usr/local/cuda/include/cudnn*.h /usr/local/cuda/lib64/libcudnn* [root@localhost cudnn-linux-x86_64-8.3.3.40_cuda10.2-archive]# cat /usr/local/cuda/include/cudnn*.h | grep CUDNN_MAJOR -A 2 四：nvidia-docker2安装 [root@localhost home]# mv /etc/yum.repos.d/openEuler.repo /etc/yum.repos.d/openEuler.repo.bak [root@localhost home]# mv repo/local.repo /etc/yum.repos.d/ [root@localhost home]# mkdir /opt/nvidia-docker2 [root@localhost home]# cp -R nvidia-docker2/* /opt/nvidia-docker2 [root@localhost home]# yum makecache [root@localhost home]# yum install nvidia-docker2 [root@localhost home]# systemctl start docker docker pull nvidia/cuda:10.2-cudnn8-runtime-centos7 nvidia-docker pull registry.baidubce.com/paddlepaddle/paddle:2.4.0-gpu-cuda10.2-cudnn7.6-trt7.0 nvidia-docker run --name paddle -it -v $PWD:/paddle registry.baidubce.com/paddlepaddle/paddle:2.4.0-gpu-cuda10.2-cudnn7.6-trt7.0 /bin/bash nvidia-sml 五：安装k8s-1.17.5 [root@localhost ~]# setenforce 0 && sed -i "s/SELINUX=enforcing/SELINUX=disabled/g" /etc/selinux/config [root@localhost ~]# systemctl stop firewalld [root@localhost ~]# systemctl disable firewalld [root@localhost ~]# echo k8s01 > /etc/hostname [root@localhost ~]# echo '192.168.83.100 k8s01' > /etc/hosts [root@localhost ~]# swapoff -a && sed -i '/ swap / s/^\(.*\)$/#\1/g' /etc/fstab [root@localhost ~]# cat <<EOF > /etc/sysctl.d/k8s.conf net.bridge.bridge-nf-call-ip6tables = 1 net.bridge.bridge-nf-call-iptables = 1 EOF [root@localhost ~]# sysctl --system [root@localhost ~]# modprobe br_netfilter [root@localhost ~]# lsmod | grep br_netfilter [root@localhost ~]# yum install yum-utils device-mapper-persistent-data lvm2 [root@localhost yum.repos.d]# systemctl start docker [root@localhost yum.repos.d]# cat <<EOF > /etc/docker/daemon.json { "registry-mirrors": [ "https://docker.mirrors.ustc.edu.cn" ], "insecure-registries": ["192.168.1.10:5000"], "exec-opts": ["native.cgroupdriver=systemd"] } EOF [root@localhost yum.repos.d]# systemctl restart docker [root@localhost yum.repos.d]# cat <<EOF > /etc/yum.repos.d/kubernetes.repo [kubernetes] name=Kubernetes baseurl=https://mirrors.aliyun.com/kubernetes/yum/repos/kubernetes-el7-x86_64/ enabled=1 gpgcheck=1 repo_gpgcheck=1 gpgkey=https://mirrors.aliyun.com/kubernetes/yum/doc/yum-key.gpg https://mirrors.aliyun.com/kubernetes/yum/doc/rpm-package-key.gpg EOF [root@localhost yum.repos.d]# yum install -y kubelet-1.22.8 kubeadm-1.22.8 kubectl-1.22.8 [root@localhost yum.repos.d]# systemctl enable kubelet.service [root@localhost yum.repos.d]# kubeadm init --kubernetes-version=1.22.8 --image-repository registry.aliyuncs.com/google_containers --apiserver-advertise-address=192.168.83.100 --service-cidr=10.1.0.0/16 --pod-network-cidr=10.244.0.0/16 --v=5 [root@localhost yum.repos.d]# systemctl enable kubelet && systemctl start kubelet [root@localhost yum.repos.d]# mkdir -p $HOME/.kube [root@localhost yum.repos.d]# sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config [root@localhost yum.repos.d]# sudo chown $(id -u):$(id -g) $HOME/.kube/config [root@localhost yum.repos.d]# kubectl apply -f https://docs.projectcalico.org/manifests/calico.yaml [root@localhost yum.repos.d]# kubectl taint nodes --all node-role.kubernetes.io/master- [root@localhost yum.repos.d]# kubectl get nodes [root@localhost yum.repos.d]# kubectl get pods -n kube-system [root@localhost yum.repos.d]# watch kubectl get pod -n kube-system -o wide kubectl get pods -n kube-system kubectl logs -f coredns-5644d7b6d9-nw2x2 -n kube-system kubectl edit deployment coredns -n kube-system