记一次aspnetcore发布部署流程初次使用k8s

主题：

aspnetcorewebapi项目，提交到gitlab，通过jenkins(gitlab的ci/cd)编译、发布、推送到k8s。

关于gitlab、jenkins、k8s安装，都是使用docker启动服务。

首先新建一个项目,为了方便浏览就把swaggerr非开发环境不展示去掉

下面就是需要准备Dockerfile和k8s.yaml文件，这里不应该用net5，过时了。

FROM mcr.microsoft.com/dotnet/aspnet:5.0-buster-slim AS base
COPY . /app
WORKDIR /app
EXPOSE 5000/tcp
ENV ASPNETCORE_URLS http://*:5000/
ENV TZ=Asia/Shanghai

# Work around for broken dotnet restore
ADD http://ftp.us.debian.org/debian/pool/main/c/ca-certificates/ca-certificates_20210119_all.deb .
RUN dpkg -i ca-certificates_20210119_all.deb

# soft link
RUN ln -snf /usr/share/zoneinfo/$TZ /etc/localtime && echo $TZ > /etc/timezone
RUN ln -s /lib/x86_64-linux-gnu/libdl-2.24.so /lib/x86_64-linux-gnu/libdl.so

# install System.Drawing native dependencies
RUN apt-get update \
    && apt-get install -y --allow-unauthenticated \
    ca-certificates \
    && update-ca-certificates \
                libgdiplus \
     && rm -rf /var/lib/apt/lists/*
     
ENTRYPOINT ["dotnet", "autopubtest.dll"]

apiVersion: apps/v1
kind: Deployment
metadata:
  name: {deployName}
  labels:
    app: {deployName}
  namespace: default
spec:
  replicas: 2
  selector:
    matchLabels:
      app: {deployName}
  template:
    metadata:
      labels:
        app: {deployName}
    spec:
      nodeSelector:
        group: web
      containers:
      - name: {containerName}
        image: {imageRegistry}/{imageName}:{imageTag}
        volumeMounts:
        - name: config-volume
          mountPath: /app/appsettings.Production.json
          subPath: appsettings.Production.json
        env:
        - name: ASPNETCORE_ENVIRONMENT
          value: Production
        ports:
        - containerPort: 5000
      volumes:
      - name: config-volume
        configMap:
          name: autopubtest-config
      imagePullSecrets:
      - name: docker-secret
---

kind: Service
apiVersion: v1
metadata:
  name:  {serviceName}
  labels:
    app:  {serviceName}
  namespace: default
spec:
  selector:
    app:  {deployName}
  ports:
  - name:  {serviceName}
    port: 5000
    protocol: TCP
    targetPort: 5000

这里需要注意的是configMap的name是我们需要再K8S里面建的appsettings.环境.json文件

  configMap:
          name: autopubtest-config

一切准备就绪，本地需要有docker环境，就能验证dockerfile是否有报错，我本地是dockerdesktop。

下面就先把代码提交到gitlab，我是用develop自建分支，而且我用的是http

这里gitlab

v17.1.1

有一个问题就是默认会把容器的id当成请求的ip地址，通过git 的git或者http拉取代码这里都会有问题，进入gitlab的容器内部找到 /etc/gitlab/gitlab.rb找到external_url注释掉的一行，改下你实际的地址和端口就行。

这里稍微提一下gitlab的ci/cd，本篇主要是jenkins。

gitlab安装完默认密码存放在 /etc/gitlab/initial_root_password ，默认用户root

networks:指定唯一，在服务器中新建一个networks，方便一个网段通信，如果是分开的服务器就是用ip或者其他。
register runner的时候手敲，ip指定gitlab容器的内网ip，查看命令 docker inspect docker容器id，类似这样的，下面提示就是成功注册一个runner
Registering runner... succeeded                     runner=

gitlab-runner register \
  --url http://gitlab的docker的ip \
  --registration-token gitlab runners中的token \
  --executor docker \
  --description "My Docker Runner" \
  --docker-image "alpine:latest"

这里是安装gitlab和gitlab-runner的docker-compose.yml 

version: '3.3'
services:
  gitlab:
    image: gitlab/gitlab-ce:latest
    container_name: gitlab
    ports:
      - "80:80"
    networks:
      - my-network

  gitlab-runner:
    image: gitlab/gitlab-runner:latest
    container_name: gitlab-runner
    restart: always
    volumes:
      - /var/run/docker.sock:/var/run/docker.sock
    networks:
      - my-network

networks:
  my-network:
    driver: bridge

只要在项目中新增.gitlab-ci.yml,再把类似jenkins的shell操作放到文件中就可以了。这里有一个测试的文件，tags很重要，注册runner的时候指定需要的，再在文件中配置了，就会按照流程。

stages:          # List of stages for jobs, and their order of execution
  - build
  - test
  - deploy

build-job:       # This job runs in the build stage, which runs first.
  stage: build
  tags:           # Add the tags here
    - docker
    - linux
  script:
    - echo "Compiling the code..."
    - echo "Compile complete."

unit-test-job:   # This job runs in the test stage.
  stage: test    # It only starts when the job in the build stage completes successfully.
  tags:           # Add the tags here
    - docker
    - linux
  script:
    - echo "Running unit tests... This will take about 60 seconds."
    - sleep 60
    - echo "Code coverage is 90%"

lint-test-job:   # This job also runs in the test stage.
  stage: test    # It can run at the same time as unit-test-job (in parallel).
  tags:           # Add the tags here
    - docker
    - linux
    - fast
  script:
    - echo "Linting code... This will take about 10 seconds."
    - sleep 10
    - echo "No lint issues found."

deploy-job:      # This job runs in the deploy stage.
  stage: deploy  # It only runs when *both* jobs in the test stage complete successfully.
  tags:           # Add the tags here
    - docker
    - linux
    - fast
  environment: production
  script:
    - echo "Deploying application..."
    - echo "Application successfully deployed."

上面仅仅只是一个测试完整流程文件，不涉及docker打包操作,需要docker打包的话runner就需要安装，安装模式有几种，自行查资料。

下面介绍jenkins的操作

这里提一提，通过git拉取代码，需要在jenkins的容器内部生成.ssh的公钥私钥，公钥添加到gitlab的ssh中，私钥就放到jenkins的全局变量中，Credentials就可以选择你的验证方式了。

下面的选择会影响你拉取代码，第一个设置你有可能需要在jenkins容器内部拉取一次代码，最后一个设置可以通过http拉。

下面继续：

这里我有三个步骤，编译，发布，K8S拉取镜像

第一部分

#!/bin/bash

echo "==========  当前 Branch: $GIT_BRANCH  =========="
echo "==========  Commit Hash: $GIT_COMMIT  =========="

cd $WORKSPACE/src/autopubtest/
  dotnet restore 

if [ -d $WORKSPACE/publish ]; then
    rm -rf $WORKSPACE/publish
fi

dotnet publish -c Release -o $WORKSPACE/publish --no-restore

if [ $? -ne 0 ]; then
  echo "!!!!!!!!!!编译失败!!!!!!!!!!"
  exit 1
else
  echo "<<<<<<<<<<编译成功>>>>>>>>>>"
fi


第二部分

#!/bin/bash
ob=`echo $jobName|tr 'A-Z' 'a-z'|cut -d '.' -f 2`
imageName="autopubtest-api"
#imageTag=$TagName
imageTag=`echo "$GIT_COMMIT" | cut -b1-8`
pushRegistry=镜像仓库/项目名
    
echo "==========  开始构建镜像 =========="

docker login -u 仓库账号 -p 仓库密码 $pushRegistry
cd $WORKSPACE/publish
docker build --rm -t $imageName:$imageTag -f $WORKSPACE/docker/Dockerfile .

if [ $? -ne 0 ]; then
  echo "!!!!!!!!!!镜像构建失败!!!!!!!!!!"
  exit 1
else
  echo "<<<<<<<<<<镜像构建成功 $imageName:$imageTag>>>>>>>>>>"
fi

                                        
docker tag $imageName:$imageTag $pushRegistry/$imageName:$imageTag
docker push $pushRegistry/$imageName:$imageTag

if [ $? -ne 0 ]; then
  echo "!!!!!!!!!!镜像发布失败!!!!!!!!!!"
  exit 1
else
  echo "<<<<<<<<<<镜像发布成功 $imageName:$imageTag>>>>>>>>>>"
fi

docker rmi $imageName:$imageTag
docker rmi $pushRegistry/$imageName:$imageTag


第三部分

projectName="autopubtest-api"
#imageTag=$TagName
imageTag=`echo "$GIT_COMMIT" | cut -b1-8`
deployName=$projectName
serviceName=$projectName
containerName=$projectName
imageName=$projectName
git_message=`git log --format=format:%s -1 ${GIT_COMMIT}`
pullRegistry=仓库地址/项目名

cat $WORKSPACE/docker/k8s.yaml | sed 's|'extensions/v1beta1'|'apps/v1'|g; s|{imageRegistry}|'$pullRegistry'|g; s|{imageName}|'$imageName'|g; s|{imageTag}|'$imageTag'|g; s|{deployName}|'$deployName'|g; s|{serviceName}|'$serviceName'|g; s|{containerName}|'$containerName'|g' > $WORKSPACE/docker/k8s.value
sed -i '/^---/,$d' $WORKSPACE/docker/k8s.value

kubectl      apply -f $WORKSPACE/docker/k8s.value 
if [ $? -ne 0 ]; then
  echo "!!!!!!!!!!更新失败，Deployment $deployName 可能不存在，尝试创建该Deployment!!!!!!!!!!"
  kubectl      create -f $WORKSPACE/docker/k8s.value 
fi

构建的日志就略过，这里使用的是harbor仓库，注需要注意，docker login需要登陆harbor的仓库，在harbor主机host通过ip地址映射一个随意取名的域名，不要用ip，否则触发https安全检查。

jenkins的第三步，会触发k8s去pull仓库镜像。关于jenkins和k8s的关联就是把k8s主机的config文件拷贝到jenkins的 ./var/jenkins_home/root/.kube/config

当K8S拉取镜像后，服务正常启动。

配置字典里新建autopubtest的appsettings.Production.json文件，该名称需要与k8s。yaml的对应起来autopubtest-config

新建下面的服务

下面就能正常使用接口了