런치챗의 배포 전략: 속도에서 안정성으로

name: LunchChat CI/CD Pipeline

on:
  push:
    branches: [main, 'dev/**']

jobs:
  build-and-push:
    runs-on: ubuntu-latest
    steps:
      - name: Build application
        run: ./gradlew build -x test --no-daemon --parallel
      
      - name: Build and push Docker image
        uses: docker/build-push-action@v5
        with:
          context: .
          push: true
          platforms: linux/amd64
          tags: ${{ env.DOCKER_REGISTRY }}/${{ env.DOCKER_IMAGE_NAME }}:${{ steps.set-tag.outputs.IMAGE_TAG }}

  # 모니터링 스택 배포
  deploy-monitoring:
    needs: build-and-push
    runs-on: ubuntu-latest
    steps:
      - name: Deploy Monitoring Stack
        uses: appleboy/ssh-action@v1.0.0
        with:
          script: |
            # 모니터링 스택 배포 로직

  # 서버1과 서버2를 병렬로 배포
  deploy-server1:
    needs: [build-and-push, deploy-monitoring]
    runs-on: ubuntu-latest
    steps:
      - name: Deploy to Server-1
        uses: appleboy/ssh-action@v1.0.0
        with:
          script: |
            export SERVER_NAME="Server-1"
            export INCLUDE_DB_MIGRATION="true"
            # ... 배포 스크립트

  deploy-server2:
    needs: [build-and-push, deploy-monitoring]  # 병렬로 실행!
    runs-on: ubuntu-latest
    steps:
      - name: Deploy to Server-2
        uses: appleboy/ssh-action@v1.0.0
        with:
          script: |
            export SERVER_NAME="Server-2"
            export INCLUDE_DB_MIGRATION="false"
            # ... 배포 스크립트

deploy-server1:
  needs: [build-and-push, deploy-monitoring]
  runs-on: ubuntu-latest
  steps:
    - name: Deploy to Server-1
      script: |
        export SERVER_NAME="Server-1"
        export INCLUDE_DB_MIGRATION="true"
        
        # 헬스체크 함수
        health_check() {
          local max_attempts=25
          echo "$SERVER_NAME 헬스체크 시작..."
          
          for i in $(seq 1 $max_attempts); do
            if curl -sf -m 5 http://localhost:8080/actuator/health | grep -q '"status":"UP"'; then
              echo "$SERVER_NAME 헬스체크 성공"
              return 0
            fi
            [ $i -eq $max_attempts ] && return 1
            sleep 12
          done
        }

verify-server1:
  needs: deploy-server1
  runs-on: ubuntu-latest
  steps:
    - name: Verify Server-1
      # 첫 번째 서버 검증

deploy-server2:
  needs: [build-and-push, verify-server1]  # 첫 번째 서버 검증 후 실행
  runs-on: ubuntu-latest
  steps:
    - name: Deploy to Server-2
      script: |
        export INCLUDE_DB_MIGRATION="false"  # 이미 첫 번째 서버에서 완료
        # ... 배포 로직

name: LunchChat CI/CD Pipeline with AWX

jobs:
  build-and-push:
    runs-on: ubuntu-latest
    outputs:
      image-tag: ${{ steps.set-tag.outputs.IMAGE_TAG }}
    steps:
      - name: Build application with Gradle
        run: ./gradlew build -x test
        
      - name: Build and push Docker image
        uses: docker/build-push-action@v5
        with:
          tags: |
            ${{ env.DOCKER_REGISTRY }}/${{ env.DOCKER_IMAGE_NAME }}:latest
            ${{ env.DOCKER_REGISTRY }}/${{ env.DOCKER_IMAGE_NAME }}:${{ github.sha }}

  deploy-app-servers:
    needs: [build-and-push, deploy-monitoring]
    runs-on: ubuntu-latest
    steps:
      - name: Install and configure AWX CLI
        run: |
          pip install awxkit
          export TOWER_HOST=${{ secrets.AWX_SERVER_URL }}
          export TOWER_TOKEN=${{ secrets.AWX_TOKEN }}
          awx config --tower_verify_ssl false
          
          EXTRA_VARS='{
            "image_tag": "${{ needs.build-and-push.outputs.image-tag }}",
            "docker_registry": "${{ env.DOCKER_REGISTRY }}",
            "docker_image_name": "${{ env.DOCKER_IMAGE_NAME }}"
          }'
          
          awx job_templates launch "Deploy LunchChat App" \
            --extra-vars "$EXTRA_VARS" \
            --limit "server1" --monitor

- name: Deploy LunchChat Application
  hosts: app_server
  become: yes
  gather_facts: no
  serial: 1  
# 롤링 배포의 핵심!

  vars:
    app_dir: "/home/ubuntu/lunchchat-app"
  
  tasks:
    - name: Create docker-compose.yml from template
      template:
        src: templates/docker-compose.app.yml.j2
        dest: "{{ app_dir }}/docker-compose.yml"
    
    - name: Pull the new image and start services
      community.docker.docker_compose_v2:
        project_src: "{{ app_dir }}"
        pull: "always"
        state: present
        recreate: always
    
    - name: Wait for container to start
      pause:
        seconds: 55
    
    - name: Wait for application to be healthy
      uri:
        url: http://localhost:8080/actuator/health
        status_code: 200
        timeout: 10
      retries: 30
      delay: 20
      register: health_check
      until: health_check.status == 200
    
    - name: Clean up unused Docker images
      docker_prune:
        images: true
        images_filters:
          dangling: false
          until: "24h"

services:
  backend:
    container_name: lunchchat-backend
    image: "{{ docker_registry }}/{{ docker_image_name }}:{{ image_tag }}"
    ports:
      - "8080:8080"
    environment:
      - SPRING_PROFILES_ACTIVE=prod
      - JAVA_OPTS=-Xms512m -Xmx1024m -XX:+UseG1GC -XX:MaxGCPauseMillis=200
      - DB_URL={{ db_url }}
      - REDIS_HOST={{ redis_host }}
      # ... 기타 환경변수들
    restart: unless-stopped
    volumes:
      - ./logs:/var/log/lunchchat
      - ./firebase-service-account-key.json:/app/firebase-service-account-key.json:ro

평균 배포 시간: 약 7-8분 (372-555초)

세부 단계별 시간:
├── Build & Push: ~2분 (119초)
├── Deploy Monitoring: ~1분 (55초)
└── Deploy App Servers: ~3분 (189초)

예상 배포 시간: 약 4-5분

세부 단계별 시간:
├── Build & Push: ~2분
├── Deploy Monitoring: ~1분
└── Deploy Server1 & Server2: ~2분 (병렬)

# GitHub Actions (이전)
- 매번 새로운 러너에서 실행
- Docker 레이어 캐시 제한적
- Gradle 캐시도 불안정

# AWX + Ansible (현재)
- 동일한 서버에서 실행
- Docker 레이어 캐시 최적화
- 로컬 Gradle 캐시 활용