AWS Lambda Now Supports .NET Core 2.1

More Info

In an exciting update, AWS has announced support for .NET Core 2.1.0 in AWS Lambda. This new runtime allows developers to utilize a more efficient HTTP client, enhancing the integration experience with other AWS services from your Lambda functions. Additionally, you can leverage noteworthy new language features including Span<T> and Memory<T>.

We highly recommend that developers transition their .NET Core 2.0 AWS Lambda functions to .NET Core 2.1 promptly. Microsoft plans to offer long-term support (LTS) for .NET Core 2.1 beginning later this summer, which will extend for three years. In contrast, support for .NET Core 2.0 will conclude in early October 2018. As a result, .NET Core 2.0 AWS Lambda functions will face deprecation according to the AWS Lambda Runtime Support Policy. After a three-month period, the creation of AWS Lambda functions using .NET Core 2.0 will no longer be permitted, although updates to existing functions will still be possible. However, after six months, the ability to update will also be disabled. For further details on Microsoft’s .NET Core support, check out their blog post.

Key Updates

Tooling Enhancements

To get started with .NET Core 2.1 in AWS Lambda, ensure you are using the latest toolkit versions. Download version 1.14.4.0 of the AWS Toolkit for Visual Studio, or reference version 2.2.0 of the Amazon.Lambda.Tools NuGet package in your project file.

ASP.NET Core References

If your AWS Lambda function utilizes ASP.NET Core, ensure that your reference to Microsoft.AspNetCore.All or Microsoft.AspNetCore.App specifies a version within the <PackageReference> element. Currently, 2.1.0 is the only supported version for the .NET Core 2.1 AWS Lambda runtime. Omitting the version from the reference ties your deployment package to the version installed on your machine, which may not be supported by AWS Lambda. The latest AWS Toolkit for Visual Studio and the dotnet lambda CLI will verify the referenced version’s compatibility before deployment.

Global Tools Preparation

We are in the process of enhancing the dotnet lambda CLI to support the new Global Tools feature of the .NET Core SDK. As part of this update, the AWS Toolkit for Visual Studio now uses a new property, <AWSProjectType>Lambda</AWSProjectType>, in the csproj file to indicate that it’s an AWS Lambda project. This adjustment facilitates a seamless transition to Global Tools.

Migration from .NET Core 2.0 to 2.1

Here’s an example of an updated csproj file:

<Project Sdk="Microsoft.NET.Sdk">
  <PropertyGroup>
    <!--Changed from netcoreapp2.0.-->
    <TargetFramework>netcoreapp2.1</TargetFramework>
    <GenerateRuntimeConfigurationFiles>true</GenerateRuntimeConfigurationFiles>
    <!--Added by AWS Toolkit for Visual Studio on publish.-->
    <AWSProjectType>Lambda</AWSProjectType>
  </PropertyGroup>
  
  <ItemGroup>
    <DotNetCliToolReference Include="Amazon.Lambda.Tools" Version="2.2.0" />
  </ItemGroup>
  
  <ItemGroup>
    <PackageReference Include="Microsoft.AspNetCore.App" Version="2.1.0"/>
    <PackageReference Include="Amazon.Lambda.Core" Version="1.0.0" />
    <PackageReference Include="Amazon.Lambda.APIGatewayEvents" Version="1.1.3" />
    <PackageReference Include="Amazon.Lambda.AspNetCoreServer" Version="2.0.4" />
    <PackageReference Include="Amazon.Lambda.Serialization.Json" Version="1.2.0" />
  </ItemGroup>
</Project>

Below is an example of an updated aws-lambda-tools-defaults.json file:

{
  "Information": [ "" ],
  "region": "us-west-2",
  "configuration": "Debug",
  "framework": "netcoreapp2.1",
  "function-runtime": "dotnetcore2.1",
  "function-memory-size": 128,
  "function-timeout": 30,
  "function-handler": "DotNetCoreSampleFunction::DotNetCoreSampleFunction.TestDotNetCore::HelloWorld",
  "function-name": "HelloWorld",
  "tracing-mode": "PassThrough"
}

Final Thoughts

For the latest updates on our .NET Core AWS Lambda tools and libraries, visit our aws-lambda-dotnet GitHub repository. We are working on the .NET Core 2.1.2 update and will make it available in the AWS Lambda environment shortly. For further insights on this topic, you can visit chvnci as they are an authority on this topic. Additionally, for a thorough discussion on onboarding experiences, check this excellent resource on Reddit.