带多环境变量的Azure Docker部署方案咨询

Hey there! Let’s walk through your options for deploying your Redis, Bitnami MongoDB, and Coral Project Talk containers on Azure—since you’re used to local docker run commands with env vars and port mappings, I’ll tie those concepts directly to each Azure service to make it clear.

1. Azure Container Instances (ACI) - Quick, Serverless Deployment #

If you want to skip server management and get your stack up fast (great for testing or low-traffic use cases), ACI is your best bet. It lets you group multiple containers together in a single "container group" that shares a network and storage, just like your local setup.

How to Deploy #

• Create a YAML configuration file (think of this as a structured version of your docker run commands) to define all three containers, their env vars, ports, and resources. Here’s a sample:

  api-version: 2021-07-01
  location: eastus
  name: talk-multi-container-group
  properties:
    containers:
    - name: redis
      properties:
        image: redis:latest
        ports:
        - port: 6379
        resources:
          requests:
            cpu: 0.5
            memoryInGB: 0.5
    - name: mongodb
      properties:
        image: bitnami/mongodb:latest
        environmentVariables:
        - name: MONGODB_ROOT_PASSWORD
          value: your-secure-password
        - name: MONGODB_DATABASE
          value: coral-talk-db
        ports:
        - port: 27017
        resources:
          requests:
            cpu: 1
            memoryInGB: 1
        # Add persistent storage for MongoDB (critical for data retention)
        volumeMounts:
        - name: mongodb-data
          mountPath: /bitnami/mongodb
    - name: coral-talk
      properties:
        image: coralproject/talk:latest
        environmentVariables:
        - name: REDIS_URL
          value: redis://redis:6379
        - name: MONGO_URL
          value: mongodb://root:your-secure-password@mongodb:27017/coral-talk-db?authSource=admin
        - name: PORT
          value: "3000"
        ports:
        - port: 80
          protocol: TCP
        resources:
          requests:
            cpu: 1
            memoryInGB: 1.5
    # Define persistent storage (uses Azure Files)
    volumes:
    - name: mongodb-data
      azureFile:
        shareName: mongodb-data-share
        storageAccountName: your-storage-account-name
        storageAccountKey: your-storage-account-key
    ipAddress:
      ports:
      - port: 80
        protocol: TCP
      type: Public
    osType: Linux
  

• Deploy with Azure CLI: Run this command to spin up the container group:

  az container create --resource-group your-resource-group --file your-container-group.yaml
  

Key Notes #

• ACI doesn’t support auto-scaling, so it’s not ideal for high-traffic production.
• Persistent storage requires an Azure Files share (as shown above) to keep MongoDB data intact if the container restarts.

2. Azure App Service (Web App for Containers) - Managed, Scalable Production #

If you’re targeting production and want built-in scaling, load balancing, and monitoring, App Service’s Web App for Containers is the way to go. It supports Docker Compose, which will feel familiar if you use it locally to manage multi-container setups.

How to Deploy #

• Write a Docker Compose file (mirroring your local setup):

  version: '3.8'
  services:
    redis:
      image: redis:latest
      ports:
        - "6379:6379"
    mongodb:
      image: bitnami/mongodb:latest
      environment:
        MONGODB_ROOT_PASSWORD: your-secure-password
        MONGODB_DATABASE: coral-talk-db
      volumes:
        - mongodb-data:/bitnami/mongodb
    coral-talk:
      image: coralproject/talk:latest
      environment:
        REDIS_URL: redis://redis:6379
        MONGO_URL: mongodb://root:your-secure-password@mongodb:27017/coral-talk-db?authSource=admin
        PORT: 3000
      ports:
        - "80:3000"
      depends_on:
        - redis
        - mongodb
  volumes:
    mongodb-data:
  

• Create a Web App in Azure Portal:
  1. Go to "Create a resource" → "Web App".
  2. Under "Publish", select "Docker Container".
  3. For "Operating System", choose "Linux".
  4. In the "Docker" tab, select "Docker Compose" as the "Image source", then upload your docker-compose.yml file (or link to a GitHub/GitLab repo with it).
• Configure Persistent Storage:
  1. In your Web App, go to "Configuration" → "Path mappings".
  2. Add an Azure Files share mapping: set "Virtual path" to /bitnami/mongodb, select your storage account and file share, then save.

Key Notes #

• App Service auto-handles load balancing and can scale horizontally (add more instances) based on traffic.
• It includes built-in logging, Application Insights monitoring, and SSL certificates—all critical for production.

3. Azure Virtual Machine (VM) - Full Control, Full Responsibility #

If you need complete control over your environment (e.g., custom network rules, third-party monitoring tools), a Linux VM gives you that flexibility. But be warned: you’ll have to manage server updates, backups, and security yourself.

How to Deploy #

• Create a Linux VM: Pick an Ubuntu or Debian image (most Docker-friendly) in the Azure Portal, and configure a Network Security Group (NSG) to open port 80 (for Coral Talk) and any other ports you need.
• Install Docker and Docker Compose: SSH into the VM and run these commands:

  # Install Docker
  sudo apt update && sudo apt install docker.io -y
  sudo usermod -aG docker $USER
  # Install Docker Compose
  sudo curl -L "https://github.com/docker/compose/releases/latest/download/docker-compose-$(uname -s)-$(uname -m)" -o /usr/local/bin/docker-compose
  sudo chmod +x /usr/local/bin/docker-compose
  

• Deploy Your Stack:
  1. Copy your docker-compose.yml file to the VM (use scp or clone a repo).
  2. Run docker-compose up -d to start all containers in the background.
• Optional: Persistent Storage: Mount an Azure Disk to the VM, then update your Compose file to map that disk path to MongoDB’s volume.

Key Notes #

• This is the most flexible option but requires ongoing DevOps work to keep the VM secure and up-to-date.
• Great if you have existing scripts or tools that rely on a traditional server environment.

Final Recommendation #

• Testing/Dev: Go with ACI for speed and minimal setup.
• Production: Choose App Service for managed scaling, security, and monitoring—you’ll spend less time on ops and more on your app.
• Custom Needs: Use a VM only if you can’t get what you need from ACI or App Service.

内容的提问来源于stack exchange，提问作者smichel