如何正确配置Docker+Nginx运行Node+Mongo？本地部署问题求助

Let's walk through each of your issues and fix them one by one. I've worked through similar Docker/Nginx/Node setups many times, so here's what's going wrong and how to resolve it:

1. Node Container Can't Connect to MongoDB #

Root Causes #

• Your Node app is trying to connect to 127.0.0.1:27017—inside a Docker container, 127.0.0.1 refers to the container itself, not the Mongo container or your host machine.
• depends_on only ensures the Mongo container starts before your Node container, but it doesn't wait for the Mongo service to actually be ready to accept connections.
• You tried container names/IPs, but likely didn't update your connection string correctly in .env.

Fixes #

1. Update your .env Mongo connection string:
   Use the Docker Compose service name (mongodb) as the host—Compose automatically creates a network where service names resolve to container IPs:

   MONGO_URI=mongodb://mongodb:27017/wetaxitask
   

2. Add health checks to Mongo to ensure Node only starts once Mongo is ready:
   Update your docker-compose.yml Mongo service:

   mongodb:
     container_name: wetaxitask_mongodb
     image: mongo:latest
     restart: always
     ports:
       - 27017:27017
     volumes:
       - ./data:/data/db
     environment:
       - MONGO_INITDB_DATABASE=wetaxitask
     healthcheck:
       test: ["CMD", "mongo", "--eval", "db.adminCommand('ping')"]
       interval: 5s
       timeout: 5s
       retries: 5
   

3. Make Node wait for Mongo's health check:
   Update your Node service in docker-compose.yml:

   wetaxitask:
     container_name: wetaxitask_api_dev
     image: wetaxitask
     restart: always
     build: .
     depends_on:
       mongodb:
         condition: service_healthy
       redis:
         condition: service_started
     env_file: .env
     ports:
       - 8080:80
     links:
       - redis
       - mongodb
   

4. Add reconnection logic to your Node code:
   If using Mongoose, enable auto-reconnect:

   mongoose.connect(process.env.MONGO_URI, {
     autoReconnect: true,
     reconnectTries: Number.MAX_VALUE,
     reconnectInterval: 1000
   });
   

2. Python-Related Errors #

Root Cause #

Your Dockerfile mixes build and production environments poorly:

• You install Python in the builder stage, but then copy the entire project to the Nginx-based production container. If your npm start triggers any build steps (like compiling native modules), the production container may have an incomplete Python environment.
• Running both Node and Nginx in the same container violates Docker's "one process per container" best practice, leading to process management issues.

Fix #

Rewrite your Dockerfile to separate build and production, and use a process manager to run both services:

# Builder Stage - handles dependencies and TypeScript compilation
FROM node:12.18-alpine AS base-builder
RUN apk update && apk add --no-cache python3 make g++ libc6-compat
WORKDIR /usr/src/app
COPY tsconfig.json package*.json ./
RUN npm ci --only=production
COPY . .
RUN npm run build # Adjust to your actual build script (e.g., tsc)

# Production Stage - runs Node + Nginx via supervisord
FROM nginx:1.19.0-alpine
RUN apk add --no-cache nodejs npm supervisor
WORKDIR /usr/src/app

# Copy only necessary files from builder
COPY --from=base-builder /usr/src/app/dist ./dist
COPY --from=base-builder /usr/src/app/node_modules ./node_modules
COPY --from=base-builder /usr/src/app/package*.json ./
COPY --from=base-builder /usr/src/app/.env ./

# Replace default Nginx config
RUN rm /etc/nginx/conf.d/default.conf
COPY nginx/nginx.conf /etc/nginx/conf.d/default.conf

# Add supervisord config to manage both processes
COPY supervisord.conf /etc/supervisord.conf

EXPOSE 80 3000
CMD ["supervisord", "-c", "/etc/supervisord.conf"]

Create a supervisord.conf file in your project root:

[supervisord]
nodaemon=true

[program:node]
command=npm start
directory=/usr/src/app
stdout_logfile=/dev/stdout
stdout_logfile_maxbytes=0
stderr_logfile=/dev/stderr
stderr_logfile_maxbytes=0

[program:nginx]
command=nginx -g "daemon off;"
stdout_logfile=/dev/stdout
stdout_logfile_maxbytes=0
stderr_logfile=/dev/stderr
stderr_logfile_maxbytes=0

3. Nginx Returns Welcome Page for All Endpoints #

Root Cause #

Your Nginx config only proxies requests to /welcome to your Node service—all other requests fall back to Nginx's default static files directory (/usr/share/nginx/html), which serves the welcome page.

Fix #

Adjust your Nginx config based on your project type:

If you have a frontend SPA + API:

server {
    listen 80;
    root /usr/src/app/dist/client; # Path to your built frontend files
    index index.html index.htm;

    # Proxy API requests to Node
    location /api/ {
        proxy_pass http://127.0.0.1:3000/api/;
        proxy_set_header Host $host;
        proxy_set_header X-Real-IP $remote_addr;
        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
        proxy_set_header X-Forwarded-Proto $scheme;
    }

    # Handle SPA routing (redirects all non-static requests to index.html)
    location / {
        try_files $uri $uri/ /index.html;
    }

    error_page 500 502 503 504 /50x.html;
    location = /50x.html {
        root /usr/share/nginx/html;
    }
}

If this is a pure API service:

server {
    listen 80;

    # Proxy all requests to Node
    location / {
        proxy_pass http://127.0.0.1:3000;
        proxy_set_header Host $host;
        proxy_set_header X-Real-IP $remote_addr;
        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
        proxy_set_header X-Forwarded-Proto $scheme;
    }

    error_page 500 502 503 504 /50x.html;
    location = /50x.html {
        root /usr/share/nginx/html;
    }
}

Bonus: Better Architecture (Separate Containers) #

For long-term maintainability, split Node and Nginx into separate containers. Update your docker-compose.yml:

version: '3.8'
services:
  wetaxitask_api:
    container_name: wetaxitask_api_dev
    build: ./api # Move Node project to an /api subdirectory with its own Dockerfile
    restart: always
    depends_on:
      mongodb:
        condition: service_healthy
      redis:
        condition: service_started
    env_file: .env
    networks:
      - app-network
  mongodb:
    container_name: wetaxitask_mongodb
    image: mongo:latest
    restart: always
    ports:
      - 27017:27017
    volumes:
      - ./data:/data/db
    environment:
      - MONGO_INITDB_DATABASE=wetaxitask
    healthcheck:
      test: ["CMD", "mongo", "--eval", "db.adminCommand('ping')"]
      interval: 5s
      timeout: 5s
      retries: 5
    networks:
      - app-network
  redis:
    container_name: wetaxitask_redis
    image: redis:latest
    ports:
      - 6379:6379
    networks:
      - app-network
  nginx:
    container_name: wetaxitask_nginx
    image: nginx:1.19.0-alpine
    restart: always
    ports:
      - 8080:80
    volumes:
      - ./nginx/nginx.conf:/etc/nginx/conf.d/default.conf
      - ./client/dist:/usr/share/nginx/html # Frontend static files
    depends_on:
      - wetaxitask_api
    networks:
      - app-network

networks:
  app-network:
    driver: bridge

Then update Nginx's proxy pass to use the Node service name:

proxy_pass http://wetaxitask_api:3000/api/;

内容的提问来源于stack exchange，提问作者Jakub