Ubuntu20.04安装国内版kubeflow1.3

由于国内网络问题，Kubeflow 通常安装都是各种磕磕碰碰，以一颗为广大人民谋福利的心，这里提供中国的本地镜像版(阿里云镜像/dockerhub)的安装。

步骤一、安装Docker

Step 1: 安装必要工具

$ sudo apt-get install apt-transport-https ca-certificates software-properties-common curl

Step 2: 安装GPG证书（密钥），并且添加Docker-ce软件源信息

# 阿里云软件源（推荐）
$ curl -fsSL https://mirrors.aliyun.com/docker-ce/linux/ubuntu/gpg | sudo apt-key add -

$ sudo add-apt-repository "deb [arch=amd64] https://mirrors.aliyun.com/docker-ce/linux/ubuntu $(lsb_release -cs) stable"

Step 3: 更新并安装Docker-ce

$ sudo apt-get update

$ sudo apt-get install docker-ce

Step 4: 将当前用户添加到 docker 组，可以不用sudo运行docker

# 将当前用户添加到 docker 组
$ sudo gpasswd -a ${USER} docker

# 重新登录或者用以下命令切换到docker组
$ newgrp - docker

# 指定存储位置（可选）

$ cat << EOF >>/etc/docker/daemon.json
{
    "data-root": "/data"
}
EOF

# 不加sudo直接执行docker命令检查效果
$ docker ps

安装kubectl
$ sudo apt install snap

$ cat << EOF >>/etc/profile
export PATH=$PATH:/snap/bin
EOF

$ source /etc/profile
$ snap install kubectl --classic

# 查看kubectl版本
$ kubectl version --client

步骤二、安装kind

$ curl -Lo ./kind https://kind.sigs.k8s.io/dl/v0.11.1/kind-linux-amd64
$ sudo chmod +x ./kind

# 把kind mv到/usr/bin
$ sudo mv ./kind /usr/bin/kind

步骤三、安装kubeflow

$ git clone https://www.aiqianji.com/openoker/kubeflow-manifests.git

$ cd kubeflow-manifests

# 创建一个单节点的k8s环境
$ kind create cluster --config=kind/kind-config.yaml --name=kubeflow --image=kindest/node:v1.17.17

# 启动成功后可以看到开了一个30000端口：

$ docker ps
CONTAINER ID   IMAGE                  COMMAND                  CREATED         STATUS         PORTS                                                 NAMES
5f67af713e28   kindest/node:v1.19.1   "/usr/local/bin/entr…"   3 minutes ago   Up 3 minutes   0.0.0.0:30000->30000/tcp, 127.0.0.1:56682->6443/tcp   kubeflow-control-plane

由于 kubeflow 实验组件较多，最好准备机器的最低配置能够大于*CPU8核,内存32G*以上。

$ python install.py

# 等待镜像拉取，由于涉及的镜像比较多，要20~30分钟左右，可以通过命令查看是否就绪：

$ kubectl get pod -nkubeflow

NAME                                                        READY   STATUS    RESTARTS   AGE
admission-webhook-deployment-6fb9d65887-pzvgc               1/1     Running   0          19h
cache-deployer-deployment-7558d65bf4-jhgwg                  2/2     Running   1          3h54m
cache-server-c64c68ddf-lx7xq                                2/2     Running   0          3h54m
centraldashboard-7b7676d8bd-g2s8j                           1/1     Running   0          4h46m
jupyter-web-app-deployment-66f74586d9-scbsm                 1/1     Running   0          3h4m
katib-controller-77675c88df-mx4rh                           1/1     Running   0          19h
katib-db-manager-646695754f-z797r                           1/1     Running   0          19h
katib-mysql-5bb5bd9957-gbl5t                                1/1     Running   0          19h
katib-ui-55fd4bd6f9-r98r2                                   1/1     Running   0          19h
kfserving-controller-manager-0                              2/2     Running   0          19h
kubeflow-pipelines-profile-controller-5698bf57cf-dhtsj      1/1     Running   0          3h52m
metacontroller-0                                            1/1     Running   0          4h52m
metadata-envoy-deployment-76d65977f7-rmlzc                  1/1     Running   0          4h52m
metadata-grpc-deployment-697d9c6c67-j6dl2                   2/2     Running   3          4h52m
metadata-writer-58cdd57678-8t6gw                            2/2     Running   1          4h52m
minio-6d6784db95-tqs77                                      2/2     Running   0          4h45m
ml-pipeline-85fc99f899-plsz2