Quickstart: Use Azure Cosmos DB for MongoDB vCore with MongoDB driver for .NET

In this quickstart, you deploy a basic Azure Cosmos DB for MongoDB vCore application using .NET. Azure Cosmos DB for MongoDB vCore is a schemaless data store allowing applications to store unstructured documents in the cloud with MongoDB libraries. You learn how to create documents and perform basic tasks within your Azure Cosmos DB resource using .NET.

Library source code | Package (NuGet) | Azure Developer CLI

• Azure Developer CLI

• Docker Desktop

• An Azure subscription

  • If you don't have an Azure subscription, create a Trial before you begin.

• .NET SDK 9.0

Use the Azure Developer CLI (azd) to create an Azure Cosmos DB for MongoDB vCore cluster and deploy a containerized sample application. The sample application uses the client library to manage, create, read, and query sample data.

1. Open a terminal in an empty directory.

2. If you're not already authenticated, authenticate to the Azure Developer CLI using azd auth login. Follow the steps specified by the tool to authenticate to the CLI using your preferred Azure credentials.

   azd auth login
   

3. Use azd init to initialize the project.

   azd init --template cosmos-db-mongodb-vcore-dotnet-quickstart
   

4. During initialization, configure a unique environment name.

5. Deploy the cluster using azd up. The Bicep templates also deploy a sample web application.

   azd up
   

6. During the provisioning process, select your subscription, desired location, and target resource group. Wait for the provisioning process to complete. The process can take approximately five minutes.

7. Once the provisioning of your Azure resources is done, a URL to the running web application is included in the output.

   Deploying services (azd deploy)
   
     (✓) Done: Deploying service web
   - Endpoint: <https://[container-app-sub-domain].azurecontainerapps.io>
   
   SUCCESS: Your application was provisioned and deployed to Azure in 5 minutes 0 seconds.
   

8. Use the URL in the console to navigate to your web application in the browser. Observe the output of the running app.

The client library is available through NuGet, as the MongoDB.Driver package. For Microsoft Entra authentication, use the Azure.Identity package from the Azure SDK for .NET.

1. Open a terminal and navigate to the /src/web folder.

   cd ./src/web
   

2. If not already installed, install the MongoDB.Driver package using dotnet add package.

   dotnet add package MongoDB.Driver
   

3. If not already installed, install the Azure.Identity package.

   dotnet add package Azure.Identity
   

4. Open and review the src/api/Microsoft.Learn.AzureCosmosDBMongoDBQuickstart.Api.csproj file to validate that both package entries exist.

Import the following namespaces into your application code:

using Azure.Core;
using Azure.Identity;

using MongoDB.Driver;
using MongoDB.Driver.Authentication.Oidc;

• Authenticate the client
• Get a database
• Get a collection
• Create a document
• Get a document
• Query documents
• Delete a document

The sample code in the template uses a database named cosmicworks and collection named products. The products collection contains details such as name, category, quantity, and a unique identifier for each product. The collection uses the /category property as a shard key.

While Microsoft Entra authentication for Azure Cosmos DB for MongoDB vCore can use well known TokenCredential types, you must implement a custom token handler. This sample implementation can be used to create a MongoClient with support for standard Microsoft Entra authentication of many identity types.

1. First, create a new class in a separate file that implements IOidcCallback interface.

   using Azure.Core;
   using MongoDB.Driver.Authentication.Oidc;
   
   internal sealed class AzureIdentityTokenHandler(
       TokenCredential credential,
       string tenantId
   ) : IOidcCallback
   {
       private readonly string[] scopes = ["https://ossrdbms-aad.database.chinacloudapi.cn/.default"];
   
       public OidcAccessToken GetOidcAccessToken(OidcCallbackParameters parameters, CancellationToken cancellationToken)
       {
           AccessToken token = credential.GetToken(
               new TokenRequestContext(scopes, tenantId: tenantId),
               cancellationToken
           );
   
           return new OidcAccessToken(token.Token, token.ExpiresOn - DateTimeOffset.UtcNow);
       }
   
       public async Task<OidcAccessToken> GetOidcAccessTokenAsync(OidcCallbackParameters parameters, CancellationToken cancellationToken)
       {
           AccessToken token = await credential.GetTokenAsync(
               new TokenRequestContext(scopes, parentRequestId: null, tenantId: tenantId),
               cancellationToken
           );
   
           return new OidcAccessToken(token.Token, token.ExpiresOn - DateTimeOffset.UtcNow);
       }
   }
   

2. Create a new instance of your custom handler class passing in a new instance of the DefaultAzureCredential type and your tenant ID.

   DefaultAzureCredential credential = new();
   
   string tenantId = "<microsoft-entra-tenant-id>";
   
   AzureIdentityTokenHandler tokenHandler = new(credential, tenantId);
   

3. Build an instance of MongoUrl using the endpoint and scheme for your recently deployed Azure Cosmos DB for MongoDB vCore instance.

   string clusterName = "<azure-cosmos-db-mongodb-vcore-cluster-name>";
   
   MongoUrl url = MongoUrl.Create($"mongodb+srv://{clusterName}.global.mongocluster.cosmos.azure.com/");
   

4. Configure your MongoClient instance using known best practice configuration options for Azure Cosmos DB for MongoDB vCore and the custom IOidcCallback implementation.

   MongoClientSettings settings = MongoClientSettings.FromUrl(url);
   
   settings.UseTls = true;
   settings.RetryWrites = false;
   settings.MaxConnectionIdleTime = TimeSpan.FromMinutes(2);
   settings.Credential = MongoCredential.CreateOidcCredential(tokenHandler);
   settings.Freeze();
   
   MongoClient client = new(settings);
   

This sample creates an instance of the IMongoDatabase interface using the GetDatabase method of the MongoClient class.

IMongoDatabase database = client.GetDatabase("<database-name>");

This sample creates an instance of the generic IMongoCollection<> interface using the GetCollection<> generic method of the IMongoDatabase interface. The generic interface and method both uses a type named Product defined in another class.

IMongoCollection<Product> collection = database.GetCollection<Product>("<collection-name>");

public record Product(
    string id,
    string category,
    string name,
    int quantity,
    decimal price,
    bool clearance
);

Create a document in the collection using collection.ReplaceOneAsync<> with the generic Product type parameter. This method "upserts" the document effectively replacing it if it already exists in the collection.

Product document = new(
    id: "aaaaaaaa-0000-1111-2222-bbbbbbbbbbbb",
    category: "gear-surf-surfboards",
    name: "Yamba Surfboard",
    quantity: 12,
    price: 850.00m,
    clearance: false
);

await collection.ReplaceOneAsync<Product>(
    doc => doc.id == document.id,
    document,
    new ReplaceOptions { IsUpsert = true }
);

Perform a read operation by using both the unique identifier (id) for the documents. Use collection.FindAsync<> with the generic Product type parameter to efficiently retrieve the specific document.

Product? document = await collection.Find(
    doc => doc.id == "aaaaaaaa-0000-1111-2222-bbbbbbbbbbbb"
).SingleOrDefaultAsync();

Perform a query over multiple documents in a container using collection.AsQueryable() and language-integrated query (LINQ). This query finds all documents within a specified category (shard key).

List<Product> documents = await collection.Find(
    filter: doc => doc.category == "gear-surf-surfboards"
).ToListAsync();

foreach (Product document in documents)
{
    // Do something with each document
}

Delete a document by sending a filter for the unique identifier of the document. Use collection.DeleteOneAsync<> to asynchronously remove the document from the collection.

await collection.DeleteOneAsync(
    doc => doc.id == "aaaaaaaa-0000-1111-2222-bbbbbbbbbbbb"
);

Use the Visual Studio Code extension for Azure Cosmos DB to explore your MongoDB vCore data. You can perform core database operations including, but not limited to:

• Performing queries using a scrapbook or the query editor
• Modifying, updating, creating, and deleting documents
• Importing bulk data from other sources
• Managing databases and collections

For more information, see How-to use Visual Studio Code extension to explore Azure Cosmos DB for MongoDB vCore data.

When you no longer need the sample application or resources, remove the corresponding deployment and all resources.

azd down --force

• Node.js Quickstart
• Python Quickstart