Setup Your Customer Folder

Before configuring, you need to create your customer-specific deployment folder:

  1. Navigate to the arena-terragrunt directory: 
    cd ./utils/arena-terragrunt/
  2. Copy the airbnb folder: 
    cp -r airbnb your-customer-name

    Replace your-customer-name with your actual customer name (e.g., acme-corp, demo-2024).

  3. Navigate to your new customer folder: 
    cd your-customer-name

💡 Note: For a hybrid approach (without VSCode Online/EKS), remove the eks-cluster folder from your customer directory after copying.

All subsequent configuration and deployment commands will be run from this customer folder.

Configuration Overview

All configuration is centralized in the config.yaml file in your customer folder. This single file controls all aspects of your deployment including MongoDB Atlas, AWS/EKS, and workshop scenario settings.

💡 Note: The HCL files (root.hcl, terragrunt.hcl, atlas-cluster/terragrunt.hcl, eks-cluster/terragrunt.hcl) automatically read from this central configuration file, so you don’t need to edit them individually.

📦 Config Backup: Every time you run terragrunt apply from the customer directory, your config.yaml file is automatically backed up to S3 at s3://mongodb-arena/terragrunt/<customer-name>/config.yaml.

Required Configuration

Open config.yaml in your customer folder and update these required values:

Customer & Domain

  • customer.name: Replace "customer" with your customer name.
  • domain.email: Replace "arena@mongodb.com" with your domain contact email.

MongoDB Atlas (Required)

  • mongodb.public_key: Replace "PUBLIC_KEY" with your MongoDB Atlas public API key.
  • mongodb.private_key: Replace "PRIVATE_KEY" with your MongoDB Atlas private API key (requires Organization Project Creator privileges).
  • mongodb.project_name: Replace "PROJECT_NAME" with your desired Atlas project name.
  • mongodb.cluster_name: Modify if needed (default: "arena-cluster").
  • mongodb.cluster_region: Modify if needed (default: "US_EAST_2").
  • mongodb.instance_size: Modify if needed (default: "M30").
  • mongodb.additional_users_count: Number of additional unassigned users to create beyond those in user_list.csv (default: 0).
  • mongodb.create_indexes: Set to true to create indexes during deployment, or false to skip index creation (default: false).
  • mongodb.database_admin_password: Password for the database admin user (default: "MongoArenaAdminDummy").
  • mongodb.customer_user_password: Password for customer users (default: "MongoArenaDummy").

AWS Configuration (Required)

  • aws.profile: AWS profile to use for authentication (default: "Solution-Architects.User-979559056307").
  • aws.region: AWS region for EKS cluster (default: "us-east-2").

Workshop Scenario Configuration (for EKS deployments)

The scenario section in config.yaml controls your workshop setup:

  • repository and branch: GitHub repository and branch to use for the workshop
  • database: Enable MongoDB and/or PostgreSQL
  • llm: LLM configuration including provider, model, and proxy settings
    • provider: Options are "anthropic" or "openai" (default: "openai")
    • model: Options are "claude-3-haiku" or "gpt-5-chat" (default: "gpt-5-chat")
  • portal: Portal configuration settings
    • admin_password: Password to access the admin portal (e.g., "ArenaAdminPassword123!")
  • leaderboard: Leaderboard configuration
    • type: Set to "timed" or "score" based (default: "timed")
    • close_on: (Optional) ISO 8601 date when leaderboard freezes in UTC (e.g., "2025-11-04T16:03:00.000Z"). When set, submissions after this time will not count. Can be managed via the Admin Portal.
    • prizes: Configure prizes promotions for the workshop
      • enabled: Set to true to enable prizes features
      • where: Location or context for the prizes (e.g., "Happy Hour")
      • when: Time for the prizes (e.g., "4:30 PM")
  • instructions.sections: Define workshop sections with titles and content paths

Note: You can leave the content arrays empty ([]) for sections if you don’t want to include specific content.

Note: For hybrid deployments (without EKS/VSCode Online), skip the scenario configuration section.

Additional Atlas Configuration (Optional)

If you need more granular control over Atlas settings:

  • Navigate to customer/atlas-cluster.
  • Update user_list.csv with SA names for testing the environment (not for workshop attendees).
  • Modify variables in terragrunt.hcl if necessary (e.g., MongoDB version, auto-scaling settings).
  • By default, a new Atlas Project is created. To use an existing project instead, the project must be imported before applying.
  • If you need to invite users, uncomment mongodbatlas_project_invitation in the Terraform files. By default, no invitations are sent.

🕐 Note: The EKS cluster expires after one week by default.

Scenario Instructions Examples

Example 1: Full Instructions Configuration

This includes all workshop sections with empty content arrays, allowing flexibility to add specific content:

instructions:
  base: "navigation-guided.yml"
  sections:
    - title: "Workshop Introduction"
      content: []
    - title: "Workshop Onboarding"
      content:
        - "/guided/"
        - "/guided/vscode/"
        - "/guided/collection/"
        - "/guided/leaderboard/"
    - title: "CRUD Operations"
      content: []
    - title: "Aggregations"
      content: []
    - title: "Search"
      content: []
    - title: "Vector Search"
      content: []

⚠️ Indentation Alert: YAML is indentation-sensitive. Make sure to maintain the exact indentation shown above (2 spaces per level). Incorrect indentation will cause deployment errors.

Example 2: Targeted Instructions Configuration

This includes specific workshop content paths for a curated experience:

⚠️ Important: If your workshop includes Search or Vector Search sections (like this example), you must enable index creation in your config.yaml:

mongodb:
  create_indexes: true

This ensures that the necessary search indexes and vector search indexes are created during deployment.

instructions:
  base: "navigation-guided.yml"
  sections:
    - title: "Workshop Introduction"
      content:
        - "/"
        - "/resources/"
    - title: "Workshop Onboarding"
      content:
        - "/guided/vscode-nop/"
        - "/guided/collection/"
    - title: "CRUD Operations"
      content:
        - "/crud/1/"
        - "/crud/2/"
        - "/crud/4/"
        - "/crud/6/"
        - "/crud/7/"
    - title: "Aggregations"
      content:
        - "/pipeline/2/"
    - title: "Search"
      content:
        - "/search/1/"
    - title: "Vector Search"
      content:
        - "/vector-search/1/"

⚠️ Indentation Alert: YAML is indentation-sensitive. Make sure to maintain the exact indentation shown above (2 spaces per level). Incorrect indentation will cause deployment errors.

💡 Key Differences:

  • Full version has empty content arrays ([]) for sections, giving you flexibility to customize all sections
  • Targeted version specifies exact workshop exercises for a focused, curated workshop experience

✅ Configuration complete? Head to Deployment: Deploy and Run to deploy your environment!