K8s 实战(4) - CentOS Docker 生产环境部署 Registry, RegistryUI, Gitea, GitWebHook 服务

Compose 安装 Gitea

version: "3.7"

networks:
  gitea:
    external: false

services:
  gitea:
    image: gitea/gitea:latest
    restart: always
    environment:
      - USER_UID=1000
      - USER_GID=1000
    ports:
      - "10022:22"
      - "10080:3000"
    volumes:
      - ./gitea:/data
      - /etc/timezone:/etc/timezone:ro
      - /etc/localtime:/etc/localtime:ro
    networks:
      - gitea

PS: 主要问题在于配置 git 的访问地址(需要修改配置文件).

Compose 安装 Registry, RegistryUI

version: "3.7"

services:
  docker-hub:
    image: registry:2
    restart: always
    ports:
      - 5000:5000
    environment:
      REGISTRY_HTTP_TLS_CERTIFICATE: /certs/xxx.pem
      REGISTRY_HTTP_TLS_KEY: /certs/xxx.key
      # REGISTRY_HTTP_SECRET: xxx
      # REGISTRY_HTTP_ADDR: 0.0.0.0:443
      REGISTRY_AUTH: htpasswd
      REGISTRY_AUTH_HTPASSWD_REALM: "Registry Realm"
      REGISTRY_AUTH_HTPASSWD_PATH: /auth/htpasswd
    volumes:
      - ./auth:/auth
      - ./certs:/certs
      - ./conf:/etc/docker/registry
      - ./registry-data:/var/lib/registry
    networks:
      - registry-ui-net

  docker-ui:
    image: joxit/docker-registry-ui:latest
    restart: always
    ports:
      - 18080:80
    environment:
      DELETE_IMAGES: 1
      #URL: http://docker-hub:5000
      REGISTRY_TITLE: "Docker Registry"
      REGISTRY_URL: https://docker-hub:5000
      #REGISTRY_HTTP_HEADERS_Access-Control-Allow-Origin: "*"
    networks:
      - registry-ui-net

networks:
  registry-ui-net:

细节和注意的问题, 参考另一篇文字: 安装 Docker Registry 和 Registry UI

PS: 主要的细节问题在于 授权, https 证书.

Compose 安装 GitWebHook

其实 Git Web Hook 是自己写的一个 http 服务, 接收 Gitea 的 Web Hook 回调, 然后执行一系列 shell 命令.

这个服务的用意是简易版的 Jenkins.

version: "3.7"

services:
  anaconda:
    image: "continuumio/anaconda3"
    restart: always
    volumes:
      - .:/data
    ports:
      - "15000:15000"
    working_dir: /data
    entrypoint: "./start.sh"

.flashenv

export FLASK_APP=main.py
export FLASK_ENV=development
export FLASK_DEBUG=1

start.sh

pip install python-dotenv
python -m flask run -h 0.0.0.0 -p 15000

PS: 主要难点在于解析数据结构, 获取项目, 分支, 提交信息, 使用自动化构建.