Prerequisites - Module 5: Data Sources & Indexers¶

Required Knowledge¶

Before starting this module, you should have:

Azure AI Search Fundamentals¶

✅ Completed Module 1: Introduction & Setup
✅ Completed Module 3: Index Management
✅ Understanding of search indexes and their structure
✅ Basic knowledge of Azure AI Search REST APIs or SDKs

Azure Services Knowledge¶

✅ Basic understanding of Azure resource management
✅ Familiarity with at least one of the following:
Azure SQL Database
Azure Blob Storage
Azure Cosmos DB
Azure Table Storage

Technical Prerequisites¶

✅ Programming experience in at least one language (Python, C#, JavaScript, or REST APIs)
✅ Understanding of JSON data format
✅ Basic knowledge of database concepts and SQL (for SQL-based sources)

Required Azure Resources¶

Azure AI Search Service¶

✅ Active Azure AI Search service (Free tier is sufficient for learning)
✅ Admin API key or appropriate RBAC permissions
✅ Service endpoint URL

Data Source Requirements¶

You'll need access to at least one of the following data sources:

Option 1: Azure SQL Database¶

✅ Azure SQL Database with sample data
✅ Database connection string
✅ SQL authentication or managed identity configured
✅ Tables with primary keys (required for indexers)

Option 2: Azure Blob Storage¶

✅ Azure Storage Account
✅ Blob container with sample documents
✅ Storage account connection string or managed identity
✅ Supported file formats (PDF, Word, Excel, JSON, etc.)

Option 3: Azure Cosmos DB¶

✅ Azure Cosmos DB account
✅ Database and container with sample data
✅ Connection string or managed identity configured
✅ Documents in JSON format

Sample Data¶

If you don't have existing data, you can use these sample datasets:

For Azure SQL Database¶

-- Sample hotels table
CREATE TABLE Hotels (
    HotelId NVARCHAR(50) PRIMARY KEY,
    HotelName NVARCHAR(100),
    Description NVARCHAR(MAX),
    Category NVARCHAR(50),
    Rating FLOAT,
    Address NVARCHAR(200),
    City NVARCHAR(50),
    StateProvince NVARCHAR(50),
    PostalCode NVARCHAR(20),
    Country NVARCHAR(50)
);

For Azure Blob Storage¶

Sample PDF documents
JSON files with structured data
Office documents (Word, Excel, PowerPoint)

For Azure Cosmos DB¶

{
  "id": "1",
  "hotelName": "Sample Hotel",
  "description": "A beautiful hotel in the city center",
  "category": "Luxury",
  "rating": 4.5,
  "address": {
    "streetAddress": "123 Main St",
    "city": "Seattle",
    "stateProvince": "WA",
    "postalCode": "98101",
    "country": "USA"
  }
}

Development Environment Setup¶

Required Tools¶

✅ Code editor (Visual Studio Code recommended)
✅ REST client (Postman, VS Code REST Client, or curl)
✅ Azure CLI (optional but recommended)

SDK Installation¶

Choose your preferred programming language:

Python¶

pip install azure-search-documents
pip install azure-identity

C# (.NET)¶

dotnet add package Azure.Search.Documents
dotnet add package Azure.Identity

JavaScript/Node.js¶

npm install @azure/search-documents
npm install @azure/identity

Authentication Setup¶

Option 1: API Keys (Simpler for learning)¶

✅ Admin API key from Azure portal
✅ Query API key (for read operations)

Option 2: Managed Identity (Recommended for production)¶

✅ System-assigned or user-assigned managed identity
✅ Appropriate role assignments:
Search Service Contributor (for admin operations)
Search Index Data Contributor (for data operations)

Network Configuration¶

Firewall Settings¶

✅ Azure AI Search service accessible from your development environment
✅ Data source services accessible from Azure AI Search
✅ Appropriate firewall rules configured

Private Endpoints (Optional)¶

✅ Private endpoint configuration if using private networking
✅ DNS resolution properly configured

Verification Checklist¶

Before proceeding with the module exercises:

Azure AI Search Service¶

[ ] Can access Azure AI Search service endpoint
[ ] Can authenticate using API keys or managed identity
[ ] Can create and manage indexes

Data Source Access¶

[ ] Can connect to your chosen data source
[ ] Sample data is available and accessible
[ ] Appropriate permissions are configured

Development Environment¶

[ ] Required SDKs or tools are installed
[ ] Can make REST API calls to Azure AI Search
[ ] Code editor is configured with appropriate extensions

Troubleshooting Common Setup Issues¶

Connection Issues¶

Verify firewall settings allow access between services
Check that connection strings are correct and up-to-date
Ensure managed identity has appropriate permissions

Authentication Problems¶

Verify API keys are valid and have correct permissions
For managed identity, check role assignments
Ensure service principal has necessary access rights

Data Access Issues¶

Confirm data source contains accessible data
Verify table/container names are correct
Check that primary keys exist for SQL-based sources

Getting Help¶

If you encounter issues during setup:

Check the troubleshooting section in this module
Review Azure AI Search documentation
Verify all prerequisites are met
Test connections independently before creating indexers

Next Steps¶

Once all prerequisites are met, proceed to: - Best Practices - Learn indexer implementation guidelines - Practice & Implementation - Start building your first indexer - Code Samples - Explore practical examples