Azure / boilerplate-azurefunctions

A boilerplate project for getting started with Azure Functions Serverless Framework plugin
MIT License
19 stars 10 forks source link
azure-functions serverless serverless-framework

Serverless Boilerplate - Azure Functions

This is a quick boilerplate for getting started with the serverless-azure-functions plugin for the serverless framework.

Getting started

1. Get a Serverless Service with Azure as the Provider

  1. Recommend using Node v6.5.0
  2. Install the serverless tooling - npm i -g serverless
  3. Create boilerplate (change my-app to whatever you'd prefer): serverless install --url https://github.com/Azure/boilerplate-azurefunctions --name my-app
  4. cd my-app
  5. npm install

2. Set up credentials

We'll set up an Azure Subscription and our service principal. You can learn more in the credentials doc.

  1. Set up an Azure Subscription

    Sign up for a free account @ https://azure.com.

    Azure comes with a free trial that includes $200 of free credit.

  2. . Get the Azure CLI

    npm i -g azure-cli
  3. Login to Azure

    azure login

    This will give you a code and prompt you to visit aka.ms/devicelogin. Provide the code and then login with your Azure identity (this may happen automatically if you're already logged in). You'll then be able to access your account via the CLI.

  4. Get your subcription and tenant id

    azure account show

    Save the subcription and tenant id for later

  5. Create a service principal for a given <name> and <password> and add contributor role.

    azure ad sp create -n <name> -p <password>

    This should return an object which has the servicePrincipalNames property on it and an ObjectId. Save the Object Id and one of the names in the array and the password you provided for later. If you need to look up your service principal later, you can use azure ad sp -c <name> where <name> is the name provided originally. Note that the <name> you provided is not the name you'll provide later, it is a name in the servicePrincipalNames array.

    Then grant the SP contributor access with the ObjectId

    azure role assignment create --objectId <objectIDFromCreateStep> -o Contributor
  6. Set up environment variables

    You need to set up environment variables for your subscription id, tenant id, service principal name, and password.

    # bash
    export azureSubId='<subscriptionId>'
    export azureServicePrincipalTenantId='<tenantId>'
    export azureServicePrincipalClientId='<servicePrincipalName>'
    export azureServicePrincipalPassword='<password>'
    # PowerShell
    $env:azureSubId='<subscriptionId>'
    $env:azureServicePrincipalTenantId='<tenantId>'
    $env:azureServicePrincipalClientId='<servicePrincipalName>'
    $env:azureServicePrincipalPassword='<password>'

3. Update the config in serverless.yml

Open up your serverless.yml file and update the following information:

service: my-azure-functions-app # Name of the Azure function App you want to create

4. Deploy, test, and remove your service

  1. Deploy a Service:

    Use this when you have made changes to your Functions or you simply want to deploy all changes within your Service at the same time.

    serverless deploy
  2. Deploy the Function:

    Use this to quickly upload and overwrite your Azure function, allowing you to develop faster.

    serverless deploy function -f httpjs
  3. Invoke the Function:

    Invokes an Azure Function on Azure

    serverless invoke --path httpQueryString.json -f httpjs
  4. Stream the Function Logs:

    Open up a separate tab in your console and stream all logs for a specific Function using this command.

    serverless logs -f httpjs -t
  5. Remove the Service: (optional)

    Removes all Functions and Resources from your Azure subscription.

    serverless remove

Contributing

Please create issues in this repo for any problems or questions you find. Before sending a PR for any major changes, please create an issue to discuss.

Please follow the Microsoft Code of Conduct when working with everyone in this project.

License

MIT