.NET Workloads on AWS Lambda

In Visual Studio, go to the Extension menu, and select Manage Extensions. On the right, in the search box, type "AWS Toolkit". In the results pane, select the "AWS Toolkit for Visual Studio" extension and click Download. Follow the prompts to install the extension, you may need to restart Visual Studio.

In relation to the Lambda service, the toolkit provides two key sets of features, project templates for Lambda functions, and tooling to assist with deployment, management, and testing/invocation of Lambda functions. Along with this, the toolkit allows you to manage other AWS resources and deploy applications to other types of AWS infrastructure such as AWS Fargate, AWS App Runner, and AWS Elastic Beanstalk (they are not covered in this course).

The toolkit adds many Lambda project templates to help you create a variety of Lambda functions.

The most simple takes a string and returns an uppercase version of that string. This is a great one to get started with because of its simplicity, ease of deployment and ease of testing. In just a few minutes you can have a .NET Lambda function running on AWS, send it requests, and get responses.

On the more complex side, there is a template that provides a full .NET 6 Web Application (with web pages and API controllers). When this is deployed, the required infrastructure to make it accessible from the web will also be deployed and configured for you.

There are container project templates available if that is your preference. Note you must have Docker installed on your computer for these to work. See the section “Choosing the runtime for you”, in module 4 for why you may choose to use container images with Lambda functions.

Unit testing your Lambda functions is just as important as testing any of your other code. Most project templates include a test project that uses xUnit to test the code in the source code project. This makes it easy to get started with unit testing.

As you can tell from the above, from Visual Studio, you can also publish your functions to the AWS Lambda service. The publish functionality takes care of building a package, uploading to the Lambda service, applying permissions, creating the API Gateway endpoint, and any other necessary steps to get your code running and available for invocation. All you have to do is write the code.

Another feature of the toolkit is the AWS Explorer, which allows you to view and manage many of your AWS resources, such as DynamoDB, S3, SNS, SQS queues, EC2, and, of course, Lambda functions.

To open the AWS Explorer, go to View menu and select AWS Explorer.

In relation to Lambda functions, the AWS Explorer lets you:

• Choose a region.
• View and manage your Lambda functions.
• Invoke Lambda functions on AWS, passing in any necessary parameters, and view the response. The tooling provides sample input to simulate calls from a variety of other AWS services. This is very useful when you are building a Lambda function that is triggered by an event from some other service, e.g. a file being created on S3.
• Debug a locally running version of a Lambda function.
• Configure services that are allowed to trigger a Lambda function.
• View logs generated by Lambda functions.
  

To invoke a Lambda function from the toolkit, open the AWS Explorer, expand the Lambda section, right click the Lambda function you want to execute, select View Function.

From there, you can choose an example request from the dropdown list. These samples let you simulate calls from other AWS services such as the API Gateway, S3, SNS, etc. This lets you easily and directly test your Lambda functions that are invoked by events from other services.

Or you can use your own request.

Along with the above, all Lambda related command line tools can be used via the terminal.

If you are new to AWS, the toolkit for Visual Studio is the best way to get started. It has the most features of all the toolkits that are available, and is the easiest to debug with, and perform deployments. The AWS toolkit for Visual Studio also works with Visual Studio Community Edition, which is available for free. This, along with the free tier on AWS is a great combination for developers who are new to AWS.

To install the AWS Toolkit for Rider, open the File menu, then Settings, and Plugins. Search for AWS Toolkit and install.

If you have already configured an AWS account on your computer, the toolkit will automatically connect to your AWS account.

Once installed, you will have a range of new AWS Lambda function templates available to you.

The AWS CLI is available for Windows, Linux, and Mac. Your method for installation will vary depending on your operating system. Full instructions can be found on the AWS page - Installing or updating the latest version of the AWS CLI. AWS advises against installing the AWS CLI using any package management software. Instead, use the link provided above.

Keep in mind that you need to update the AWS CLI from time to time to get the latest features.

To check your version of the AWS CLI, run the following command:

You will see a response that looks similar to this:

This tells me I have version 2.5.4 of the AWS CLI installed.

To see what is the most recent version of the AWS CLI, go to https://github.com/aws/aws-cli/blob/v2/CHANGELOG.rst, below “CHANGELOG” will be the version of the most recent release. At the time of this writing it is 2.7.2, so I need to update my AWS CLI!

With the AWS CLI you can manage all your AWS resources. The AWS CLI includes powerful querying and filtering capabilities, which can help you find the resources you are looking for. As you would expect, the AWS CLI allows you to list, view, manage, and invoke Lambda functions.

Extensive documentation with examples is available at the AWS CLI Command Reference.

To install the templates run the following command:

After the installation is complete, you will use the dotnet new command to create new Lambda function projects.

To see the list of project templates available, run the following command:

You will see output that looks like this (note the available templates change periodically):

You might notice that the short name for some templates starts with "serverless." and some start with "lambda.".

Those that start with "serverless." use the AWS Serverless Application Model (SAM) to deploy the Lambda function and build the required infrastructure that the Lambda function will use. AWS SAM is an open source framework that let’s you build serverless application on AWS. Along with defining the Lambda function, with AWS SAM you can define other resources such as API Gateways, databases, event sources.

If you navigate to the projectname/src/projectname directory, you will see a file called serverless.template. This defines the required infrastructure, and how it is wired up to the Lambda function. In all the "serverless." templates listed above, an API Gateway is used to route HTTP requests to the Lambda function.

The templates that start with "lambda." do not use the Serverless Application Model, and do not setup other infrastructure. If other infrastructure is required, you will need to set it up yourself. For example, the lambda.Kinesis project template shows you how to process incoming Kinesis events, but it does NOT create the Kinesis streams for you, or add the relevant permissions needed for Kinesis to trigger the Lambda function.

To create a new project based on the lambda.EmptyFunction template run:

The tooling is open source and available to view on GitHub,

https://github.com/aws/aws-lambda-dotnet#dotnet-cli-templates.

As with the other command line tools, you should keep an eye on the relevant GitHub pages to see when new versions are released.

To install the AWS Extension for .NET CLI, run the following:

To verify it is installed correctly, run:

This will output a summary of commands you can use and a version number.

This tool works from any shell and is especially useful when used from the terminal within Visual Studio Code and Rider. It allows you to perform common Lambda function related tasks.

For example, to list all Lambda functions, you can use:

Or to invoke a Lambda function, run:

It can also deploy Lambda functions. After creating a Lambda function, with dotnet Lambda templates, (see above) you can deploy the function to AWS using -

This will kick off a build and guided deployment process. The tool will ask you for the Lambda function name, the role to assign to the function (or create a new one). If you created a new role, it will ask you to choose a permission policy for that new role. Then the tool will deploy the function to AWS and within a few seconds you'll be able to invoke it!

To see all available command line options, use -

Keep in mind that behind each command there can be a lot of functionality, so it is important to understand that each command line supports multiple options. For example, the update-function-config is where you enable and configure Lambda function URLs. That functionality is not available in the deploy-function command. Keeping up with the latest release documentation and features on the Lambda service is important, as it will help you understand and make use of the latest features.

While this tool is very convenient and powerful for managing Lambda functions when developing code and learning about AWS, we strongly encourage you to use a robust CI/CD process for deploying to your applications when they reach a certain level of maturity. AWS supports all major CI/CD tools.

For installation on Windows, see the Installing the AWS Tools for PowerShell on Windows guide.

For installation on Linux or Mac, see the Installing AWS Tools for PowerShell on Linux or macOS guide.

The number of modules, and then commands that can be performed by AWS Tools for PowerShell is very large. In relation AWS Lambda, you can use the tools to list, view, deploy, manage, and invoke functions.

To see all available commands run:

You can find full documentation for all commands in the AWS Tools for PowerShell Cmdlet Reference, navigate to the Lambda section in the left bar.

To list all Lambda functions run:

To invoke a Lambda function run: