当前位置：首页 > news >正文

Jenkins运维之路(Jenkins流水线改造Day02-3-容器项目)

news 2025/9/17 6:27:19

上期已经将容器项目的告警补充完成，这期在补充下ansible的相关部署内容，项目我使用的是自建的运维测试项目。

1. Ansible 目录结构

# 代码目录结构
tree -L 3
.
├── Ansible # Ansible部署脚本目录
│   ├── HelloWorld-Pipeline  # 按Jenkins项目名建立playbook文件夹
│   │   ├── deploy.yml  # 部署脚本
│   │   ├── hosts # 主机清单
│   │   └── rollback.yml # 回滚脚本
│   └── Readme.md
├── Jenkinsfile
│   ├── Helloworld-Pipeline-DockerSlave.jenkinsfile
│   ├── HelloWorld-Pipeline.jenkinsfile
│   └── Readme.md
├── Readme.md
├── src
│   └── org
│       └── foo
└── vars└── Readme.md

2. Ansible部署脚本(deploy.yml)

---
- name: Deploy Docker containerhosts: testbecome: yesvars:# 这些变量都会从Jenkins中传入new_image_name: "{{ new_image_name }}"container_name: "{{ container_name }}"docker_registry: "{{ docker_registry }}"  docker_username: "{{ docker_username }}" docker_password: "{{ docker_password }}"tasks:- name: Login to Docker registrydocker_login:registry: "{{ docker_registry }}"username: "{{ docker_username }}"password: "{{ docker_password }}"- name: Pull the latest Docker imagedocker_image:name: "{{ new_image_name }}"tag: latestsource: pull- name: Stop and remove old container if existsdocker_container:name: "{{ container_name }}"state: absentignore_errors: yes  # 忽略错误，确保后续任务继续执行- name: Run new containerdocker_container:name: "{{ container_name }}"image: "{{ new_image_name }}"state: startedrestart_policy: alwayspublished_ports:- "8888:8080"  # 映射主机的8888端口到容器的8080端口- name: Remove unused Docker imagescommand: docker image prune -f --filter "until=24h"  # 清理24小时内未使用的镜像，这个根据自己情况register: prune_result

3. Ansible部署脚本(rollback.yml)

---
- name: Deploy Docker containerhosts: testbecome: yesvars:# 这些变量都会从Jenkins中传入new_image_name: "{{ rel_image_name }}"container_name: "{{ container_name }}"docker_registry: "{{ docker_registry }}"docker_username: "{{ docker_username }}"docker_password: "{{ docker_password }}"ansible_python_interpreter: /usr/bin/python3tasks:- name: Login to Docker registrydocker_login:registry: "{{ docker_registry }}"username: "{{ docker_username }}"password: "{{ docker_password }}"- name: Pull the latest Docker imagedocker_image:name: "{{ rel_image_name }}"tag: latestsource: pull- name: Stop and remove old container if existsdocker_container:name: "{{ container_name }}"state: absentignore_errors: yes- name: Run new containerdocker_container:name: "{{ container_name }}"image: "{{ rel_image_name }}"state: startedrestart_policy: alwayspublished_ports:- "8888:8080"- name: Remove unused Docker imagescommand: docker image prune -f --filter "until=24h"register: prune_result

4. Hosts

[test]
192.168.1.98

5.Jenkins传入变量的代码片段(在前面2片文章有写)

pipeline {//.......省略stages {//.......省略//这里开始传入相关变量stage('Ansible Deploy') {steps {ansiColor('xterm') { // 启用 AnsiColorecho "${PURPLE}Ansible 部署项目${RESET}"}script {sh "git clone ${OPS_SHARE_LIBRARY} ./ops-share-librarya"withCredentials([usernamePassword(credentialsId: 'Harbor', usernameVariable: 'HARBOR_USERNAME', passwordVariable: 'HARBOR_PASSWORD')]) {def extraVars = [container_name: "${CONTAINER_NAME}",docker_registry: "${HARBOR_URL}",docker_username: "${HARBOR_USERNAME}",docker_password: "${HARBOR_PASSWORD}"].collectEntries { [(it.key): it.value] }// 根据 BRANCH_TAG 的值添加特定的参数if (!env.BRANCH_TAG.startsWith('rel-')) {// 正常部署时添加的参数extraVars['new_image_name'] = "${REGISTRY_URL}/${PROJECT_GROUP}/${PROJECT_NAME}:${_tag}"// 执行正常的部署ansiblePlaybook(playbook: "./ops-share-librarya/Ansible/HelloWorld-Pipeline/deploy.yml",inventory: "./ops-share-librarya/Ansible/HelloWorld-Pipeline/hosts",extraVars: extraVars)} else {def branchTag = env.BRANCH_TAGdef extractedValue = branchTag.replaceFirst(/^rel-/, '')  // 去掉前缀 "rel-"println "Extracted Value: ${extractedValue}"  // 输出提取的值// 回滚时添加的参数extraVars['rel_image_name'] = "${REGISTRY_URL}/${PROJECT_GROUP}/${PROJECT_NAME}:${extractedValue}"// 执行回滚ansiblePlaybook(playbook: "./ops-share-librarya/Ansible/HelloWorld-Pipeline/rollback.yml",inventory: "./ops-share-librarya/Ansible/HelloWorld-Pipeline/hosts",extraVars: extraVars)}}}}}//.......省略}
}