欢迎您访问 最编程 本站为您分享编程语言代码，编程技术文章！

热门搜索/Hot Search

您现在的位置是： 首页

GitLab持续集成/部署（CI/CD）与GitLab Runner、Docker的实战指南与经验分享

最编程 2024-02-22 09:18:32
...

GitLab CI/CD

Docker安装gitlab-runner

$ docker run -d --name gitlab-runner --restart always \
     -v /srv/gitlab-runner/config:/etc/gitlab-runner \
     -v /var/run/docker.sock:/var/run/docker.sock \
     gitlab/gitlab-runner:latest

可以在运行runner时暴露一个端口，用于在gitlab流水线上直接debug：

$ docker run -d --name gitlab-runner -p 8081:8081 --restart always \
     -v /srv/gitlab-runner/config:/etc/gitlab-runner \
     -v /var/run/docker.sock:/var/run/docker.sock \
     gitlab/gitlab-runner:latest

运行后需要修改runner的配置文件/srv/gitlab-runner/config/config.toml：

[session_server]
	listen_address = "[::]:8081"
	advertise_address = "IP:8081"
	session_timeout = 1800
[runners.docker]
	volumes = ["/cache","/usr/bin/docker:/usr/bin/docker","/var/run/docker.sock:/var/run/docker.sock"]
	pull_policy = "if-not-present"

进入gitlab-runner

$ docker exec -it gitlab-runner bash

查看runners的状态

$ gitlab-runner status

重启runners

$ gitlab-runner run

注册gitlab-runner

$ docker exec -it gitlab-runner gitlab-runner register

执行后需要依次输入提示信息

其中url和token可以在gitlab项目设置里面找到

image.png

.gitlab-ci.yml

  • stages

    全局自定义阶段

    stages:
	- stage1
	- stage2
	- stage3

  • script

    shell脚本

    job1:
	stage: stage1
	script: "echo 'this is job1'"
	
job2:
	stage: stage2
	script:
		- echo "this is job2"
		- echo "this is job2 too"

  • stage

    任务内的阶段，必须从全局阶段中选

    job1:
	stage: stage1
	script: "echo 'this is stage1'"

  • retry

    失败后的重试次数（0不会重试，2最多重试2次）

    when：当什么时候重试

    test:
	script: rspec
	retry: 0
test:
	script: rspec
	retry:
		max: 2
		when:
			- runner_system_failure
			- syuck_or_timeout_failure

  • image

    指定一个基础Docker镜像作为基础运行环境，经常用到的镜像有node java python docker

  • tags

    tags关键字是用于指定Runner，tags的取值范围是在该项目可见的runner tags中

  • only/except

    限定当前任务执行的条件

  • when

    when关键字是实现在发生故障或尽管发生故障仍能运行的作业

  • cache

    是将当前工作环境目录中的一些文件，一些文件夹存储起来，用于在各个任务初始化的时候恢复

  • variables

    设置变量的三种方式：

    • 在.gitlab-ci.yml中自己定义
    • 使用pipeline中预定义的变量
    • 设置在CICD中设置变量

  • interruptible

    默认为false，可以设置为true，会自动取消旧的正在运行的流水线（需要在设置中CI/CD设置Auto-cancel redundant pipelines）

  • timeout

    设置超时时间

  • rousource_group

    限定分支的部署任务数量只能有一个，其他新的流水线会进入等待状态

一个前端自动部署的例子：

image: node:alpine

stages:
	- install
	- build
	- deploy

cache:
	key: modules-cache
	paths:
		- node_modules

job_install:
	stage: install
	tags: 
	 - dockercicd
	script:
		- npm install
	interruptible: true
	rouscource: prod
		
job_build:
	stage: build
	tags: 
	 - dockercicd
	script:
		- npm run build
	only:
		- release # 只在release分支执行
	interruptible: true
	rouscource: prod
		
job_deplot:
	stage: deploy
	variables:
		WEB_NAME:"web-image"
	image: docker
	tags: 
	 - dockercicd
	script:
		- docker build -t $WEB_NAME . # 当前项目目录下写了Dockerfile 执行生成镜像 Dockerfile如下
		- if [ $(docker ps -aq --filter name=$WEB_NAME) ];then docker rm -f $WEB_NAME;fi
		- docker run -d -p 8082:80 $WEB_NAME
	when: manual # 手动执行任务
	interruptible: true
	rouscource: prod

FROM node:lastest as builder
WORKDIR /app
COPY package.json .
RUN npm install --registry=http://registry.npm.taobao.org
COPY . .
RUN npm run build

FROM nginx:lastest
COPY --from=builder /app/dist /usr/share/nginx/html

CI/CD相关设置

取消默认的邮件提醒

image-20210621130239524.png

流水线没有权限触发

Pipeline failed due to the user not being verified

image-20210621130443174.png

这个问题比较奇葩，网上找了很久才在外网找到问题所在，是因为默认启用了允许使用共享tags导致需要验证身份的问题。

这里我选择直接关闭，同样是在获取token那个页面。

流水线

  • 基本流水线

  • DAG流水线

    stages:
	- build
	- test
	- deploy

image: alpine

build_a:
	stage: build
	script:
		- echo "this job builds something quickly"

build_b:
	stage: build
	script:
		- echo "this job builds something slowly"
		
test_a:
	stage: test
	need: [build_a]
	script:
		- echo "this test job will start as soon as build_a finished"
		- echo "it will not wait for build_b,or other jobs in the build stage,to finished"
		
test_b:
	stage: test
	need: [build_b]
	script:
		- echo "this test job will start as soon as build_b finished"
		- echo "it will not wait for build_b,or other jobs in the build stage,to finished"
		
deploy_a:
	stage: deploy
	need: [test_a]
	script:
		- echo "Since build_a and test_a run quickly,this deploy job can run much earlier"
		- echo "it dose not need to wait for build_b or test_b"

  • 父子流水线

    stages:
	- test
	- triggers
	- deploy
	
testjob:
	stage: test
	script: echo 'test job'
	
trigger_a:
	stage: triggers
	trigger:
		include: a/.gitlab-ci.yml
	rules:
		- changes:
			- a/*

trigger_b:
	stage: triggers
	trigger:
		include: b/.gitlab-ci.yml
	rules:
		- changes:
			- b/*

  • 多项目流水线

    stages:
	- test
	- triggers
	- deploy
	
testjob:
	stage: test
	script: echo 'test job'
	
trigger_a:
	stage: triggers
	trigger:
		include: a/.gitlab-ci.yml
	rules:
		- changes:
			- a/*

trigger_b:
	stage: triggers
	trigger:
		include: b/.gitlab-ci.yml
	rules:
		- changes:
			- b/*
			
trigger_c:
	stage: deploy
	trigger:
		project: root/pain_html
		branch: master
		stategy: depend

  • 合并请求流水线

    image: node:alpine

stages:
	- install
	- build
	- deploy

cache:
	key: modules-cache
	paths:
		- node_modules

job_install:
	stage: install
	tags: 
	 - dockercicd
	script:
		- npm install
		
job_build:
	stage: build
	tags: 
	 - dockercicd
	script:
		- npm run build
	only:
		- merge_requests # 只在merge时执行

流水线的触发

  • 推送代码
  • 定时触发
  • url触发
  • 手动触发

vue项目实战 .gitlab-ci.yml

stages: 
  - package
  - build
  - deploy
my_package:
  image: 192.168.100.150/tools/node-git:10-alpine
  stage: package
  script:
    - npm install --registry=https://registry.npm.taobao.org
    - npm run build:prod
    - cp Dockerfile dist/Dockerfile
  cache:
    key: ${CI_PIPELINE_ID}
    paths: 
      - dist/
  only:
    - master
  tags:
   - runInDk
my_build:
  stage: build
  cache:
    key: ${CI_PIPELINE_ID}
    paths: 
      - dist/
  script:
    - cd dist
    - docker build -t 192.168.88.4:5000/${CI_PROJECT_NAME}:${CI_PIPELINE_ID} .
    - docker push 192.168.88.4:5000/${CI_PROJECT_NAME}:${CI_PIPELINE_ID}
  tags:
    - runInDk
my_deploy:
  stage: deploy
  script:
    - docker stop ${CI_PROJECT_NAME} && docker rm ${CI_PROJECT_NAME}
    - docker run -d -p 8080:80 --restart=always --name=${CI_PROJECT_NAME} 192.168.88.4:5000/${CI_PROJECT_NAME}:${CI_PIPELINE_ID}
  tags:
    - runInDk

上一篇： Blender爱好者交流平台，热烈邀请大家进群畅聊哦！

下一篇： 学习交流群

推荐阅读