#AWS - Invoke Local

This runs your code locally by emulating the AWS Lambda environment. Please keep in mind, it's not a 100% perfect emulation, there may be some differences, but it works for the vast majority of users. We mock the context with simple mock data.

serverless invoke local --function functionName

Note: Please refer to this guide for event data passing when your function uses the http event with a Lambda Proxy integration.

#Options

• --function or -f The name of the function in your service that you want to invoke locally. Required.
• --path or -p The path to a json file holding input data to be passed to the invoked function as the event. This path is relative to the root directory of the service.
• --data or -d String data to be passed as an event to your function. Keep in mind that if you pass both --path and --data, the data included in the --path file will overwrite the data you passed with the --data flag.
• --raw Pass data as a raw string even if it is JSON. If not set, JSON data are parsed and passed as an object.
• --contextPath or -x, The path to a json file holding input context to be passed to the invoked function. This path is relative to the root directory of the service.
• --context or -c, String data to be passed as a context to your function. Same like with --data, context included in --contextPath will overwrite the context you passed with --context flag.

• --env or -e String representing an environment variable to set when invoking your function, in the form <name>=<value>. Can be repeated for more than one environment variable.
• --docker Enable docker support for NodeJS/Python/Ruby/Java. Enabled by default for other runtimes.
• --docker-arg Pass additional arguments to docker run command when --docker is option used. e.g. --docker-arg '-p 9229:9229' --docker-arg '-v /var:/host_var'
• --skip-package Use the last packaged files from .serverless directory. This will speed up invocation significantly as we can skip the packaging of all files before every invoke

#Environment

The invoke local command sets reasonable environment variables for the invoked function. All AWS specific variables are set to values that are quite similar to those found in a real "physical" AWS Lambda environment. Additionally the IS_LOCAL variable is set, that allows you to determine a local execution within your code.

#Examples

#Local function invocation

serverless invoke local --function functionName

This example will locally invoke your function.

#Local function invocation with data

serverless invoke local --function functionName --data "hello world"

serverless invoke local --function functionName --data '{"a":"bar"}'

#Local function invocation with data from standard input

node dataGenerator.js | serverless invoke local --function functionName

#Local function invocation with data passing

serverless invoke local --function functionName --path lib/data.json

This example will pass the json data in the lib/data.json file (relative to the root of the service) while invoking the specified/deployed function.

#Example data.json

{
  "resource": "/",
  "path": "/",
  "httpMethod": "GET"
  //  etc. //
}

#Local function invocation with custom context

serverless invoke local --function functionName --context "hello world"

#Local function invocation with context passing

serverless invoke local --function functionName --contextPath lib/context.json

This example will pass the json context in the lib/context.json file (relative to the root of the service) while invoking the specified/deployed function.

#Local function invocation, setting environment variables

serverless invoke local -f functionName -e VAR1=value1

# Or more than one variable

serverless invoke local -f functionName -e VAR1=value1 -e VAR2=value2

When using AWS CloudFormation intrinsic functions as environment variables value, only Fn::ImportValue and Ref will be automatically resolved for function invocation. Other intrinsic functions use will result in the corresponding configuration object passed in the function as environment variable.

functions:
  functionName:
    handler: handler.main
    environment:
      EXT_TABLE_NAME:
        Fn::ImportValue: exported-tableName
      REF_TABLE_NAME:
        Ref: myTable
      INT_TABLE_NAME:
        Fn::GetAtt: [myTable, Arn]

In the above example, EXT_TABLE_NAME and REF_TABLE_NAME will be resolved to the exported value exported-tableName and myTable physical ID respectively while INT_TABLE_NAME will not be resolved.

#Limitations

Use of the --docker flag and runtimes other than NodeJs, Python, Java, & Ruby depend on having Docker installed. On MacOS & Windows, install Docker Desktop; On Linux install Docker engine and ensure your user is in the docker group so that you can invoke docker without sudo.

Note: In order to get correct output when using Java runtime, your Response class must implement toString() method.

Environment variables: The IS_LOCAL environment variable, as well as any environment variables provided via command line arguments, will only be set once the invoked function begins its execution. They will not be set during the parsing of the serverless.yml file.

#Resource permissions

Lambda functions assume an IAM role during execution: the framework creates this role, and set all the permission provided in the iamRoleStatements section of serverless.yml.

Unless you explicitly state otherwise, every call to the AWS SDK inside the lambda function is made using this role (a temporary pair of key / secret is generated and set by AWS as environment variables, AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY).

When you use serverless invoke local, the situation is quite different: the role isn't available (the function is executed on your local machine), so unless you set a different user directly in the code (or via a key pair of environment variables), the AWS SDK will use the default profile specified inside your AWS credential configuration file.

Take a look to the official AWS documentation (in this particular instance, for the javascript SDK, but should be similar for all SDKs):

• http://docs.aws.amazon.com/sdk-for-javascript/v2/developer-guide/loading-node-credentials-shared.html
• http://docs.aws.amazon.com/sdk-for-javascript/v2/developer-guide/loading-node-credentials-lambda.html

Whatever approach you decide to implement, be aware: the set of permissions might be (and probably is) different, so you won't have an exact simulation of the real IAM policy in place.