Skip to main content

Collection of tools to enable use of AWS Lambda with CloudFormation.

Project description

cfnlambda is a collection of AWS Lambda tools to enable use of AWS Lambda functions with CloudFormation. At it’s core it is the cfn_response function and the handler_decorator decorator. These enable an AWS Lambda function, launched from a CloudFormation stack, to log to CloudWatch, return data to the CloudFormation stack and gracefully deal with exceptions.

Quickstart

The easiest way to use cfnlambda is to use the handler_decorator decorator on your AWS Lambda function.

from cfnlambda import handler_decorator

@handler_decorator()
def lambda_handler(event, context):
    result = (float(event['ResourceProperties']['key1']) +
              float(event['ResourceProperties']['key2']))
    return {'sum': result}

handler_decorator

When you decorate your AWS Lambda function with the handler_decorator a few things happen. Your AWS Lambda function can now emit output back to the CloudFormation stack that launched it simply by returning a dictionary of key/value pairs, all of which become available to the CloudFormation stack as attributes of the custom resource in the stack. These values can then be accessed with the Fn::GetAtt CloudFormation function.

{ "Fn::GetAtt": [ "MyCustomResource", "a_key_returned_by_my_lambda_function" ] }

Any non-dictionary returned will be put into an custom resource attribute called result. Any exceptions raised by your AWS Lambda function will be caught by handler_decorator, logged to the CloudWatch logs and returned to your CloudFormation stack in the result attribute.

{ "Fn::GetAtt": [ "MyCustomResource", "result" ] }

Unless the delete_logs argument is set to False in handler_decorator, all CloudWatch logs generated while the stack was created, updated and deleted will be deleted upon a successful stack deletion. If an exception is thrown during stack deletion, the logs will always be retained to facilitate troubleshooting. To force retention of logs after a stack is deleted, set delete_logs to False.

from cfnlambda import handler_decorator
logging.getLogger().setLevel(logging.DEBUG)

@handler_decorator(delete_logs=False)
def lambda_handler(event, context):
    mirror_text = event['ResourceProperties']['key1'][::-1]
    return {'MirrorText': mirror_text}

Finally, AWS Lambda functions decorated with handler_decorator will not report a status of FAILED when a stack DELETE is attempted. This will prevent a CloudFormation stack from getting stuck in a DELETE_FAILED state. One side effect of this is that if your AWS Lambda function throws an exception while trying to process a stack deletion, though the stack will show a status of DELETE_COMPLETE, there could still be resources which your AWS Lambda function created which have not been deleted. To disable this feature, pass hide_stack_delete_failure=False as an argument to handler_decorator.

from cfnlambda import handler_decorator

@handler_decorator(hide_stack_delete_failure=False)
def lambda_handler(event, context):
    raise Exception(
        'This will result in a CloudFormation stack stuck in a
        DELETE_FAILED state')

handler_decorator usage walkthrough

Here is an example showing the creation of a very simple AWS Lambda function which sums two values passed in from the CloudFormation stack (‘key1’ and ‘key2) and returns the result back to the stack as ‘sum’.

Example assumptions:

  • You have a pre-existing s3 bucket called example-bucket-us-west-2 in the us-west-2 region which is either public or readable by the user launching the CloudFormation stack.

  • You have some way to upload a file into that s3 bucket. In the example we’re using the AWS CLI tool. Here’s how to install and configure AWS CLI.

First, this Lambda code must be zipped and uploaded to an s3 bucket.

from cfnlambda import handler_decorator
import logging
logging.getLogger().setLevel(logging.INFO)

@handler_decorator()
def lambda_handler(event, context):
    result = (float(event['ResourceProperties']['key1']) +
              float(event['ResourceProperties']['key2']))
    return {'sum': result}

Here are a set of commands to create and upload the AWS Lambda function

dir=/path/to/PythonExampleDir
mkdir $dir

# Create your AWS Lambda function
cat > $dir/example_lambda_module.py <<End-of-message
from cfnlambda import handler_decorator
import logging
logging.getLogger().setLevel(logging.INFO)

@handler_decorator()
def lambda_handler(event, context):
    result = (float(event['ResourceProperties']['key1']) +
              float(event['ResourceProperties']['key2']))
    return {'sum': result}
End-of-message

pip install cfnlambda --no-deps -t $dir
zip --junk-paths $dir/example_lambda_package.zip $dir/*
aws --region us-west-2 s3 cp $dir/example_lambda_package.zip s3://example-bucket-us-west-2/

Next, the CloudFormation template must be written. Here is an simple example CloudFormation stack that uses the Lambda function above. To use this example, save this template to a file called example_cloudformation_template.json

{
  "Resources" : {
    "SumInfo": {
      "Type": "Custom::SumInfo",
      "Properties": {
        "ServiceToken": { "Fn::GetAtt" : ["ExecuteSum", "Arn"] },
        "key1": "1.2",
        "key2": "5.9"
      }
    },
    "ExecuteSum": {
      "Type": "AWS::Lambda::Function",
      "Properties": {
        "Handler": "example_lambda_module.lambda_handler",
        "Role": { "Fn::GetAtt" : ["LambdaExecutionRole", "Arn"] },
        "Code": {
          "S3Bucket": "example-bucket-us-west-2",
          "S3Key": "example_lambda_package.zip"
        },
        "Runtime": "python2.7"
      }
    },
    "LambdaExecutionRole": {
      "Type": "AWS::IAM::Role",
      "Properties": {
        "AssumeRolePolicyDocument": {
          "Version": "2012-10-17",
          "Statement": [{
              "Effect": "Allow",
              "Principal": {"Service": ["lambda.amazonaws.com"]},
              "Action": ["sts:AssumeRole"]
          }]
        },
        "Policies": [{
          "PolicyName": "root",
          "PolicyDocument": {
            "Version": "2012-10-17",
            "Statement": [
              {
                "Effect": "Allow",
                "Action": ["logs:CreateLogGroup", "logs:CreateLogStream", "logs:PutLogEvents"],
                "Resource": "arn:aws:logs:*:*:*"
              },
              {
                "Effect": "Allow",
                "Action": ["logs:DeleteLogGroup"],
                "Resource": {"Fn::Join":["", ["arn:aws:logs:", {"Ref":"AWS::Region"},":",{"Ref":"AWS::AccountId"}, ":log-group:/aws/lambda/*"]]}
              }
            ]
          }
        }]
      }
    }
  },
  "Outputs" : {
    "Sum" : {
      "Description" : "The sum of the two values",
      "Value" : { "Fn::GetAtt": [ "SumInfo", "sum" ] }
    }
  }
}

Next, the CloudFormation template must be uploaded to execute the AWS Lambda function.

aws --region us-west-2 cloudformation create-stack --capabilities CAPABILITY_IAM --stack-name ExampleCloudFormationStack --template-body file:///home/user/example_cloudformation_template.json

Finally, you can see that the CloudFormation stack was created and the Lambda function executed by looking at the CloudWatch logs that it created or at the CloudFormation stack output. You should see in the stack output the “sum” of the “key1” and “key2”

aws --region us-west-2 cloudformation describe-stacks --stack-name ExampleCloudFormationStack

cfn_response

cfn_response is a Python function designed as a drop in replacement for the Node.js cfn-response function provided by AWS. It accepts the same arguments and does the same thing.

cfn_response allows your AWS Lambda function to communicate out to the CloudFormation stack that launched it. This communication is done through an AWS signed URL. Here’s an example of cfn_response in use

from cfnlambda import cfn_response, Status, RequestType

def lambda_handler(event, context):
    client = boto3.client('ec2')
    if event['RequestType'] == RequestType.DELETE:
        client.delete_key_pair(KeyName='example-cfnlambda-keypair')
        result = {'result': 'Key deleted'}
    else:
        keypair = client.create_key_pair(KeyName='example-cfnlambda-keypair')
        result = {'result': 'Key created',
                  'KeyMaterial': keypair['KeyMaterial']}
    cfn_response(event,
                 context,
                 Status.SUCCESS,
                 result)

This example would send the KeyMaterial (SSH private key) back to the CloudFormation stack where it could be accessed like this

{ "Fn::GetAtt": [ "MyCustomResource", "KeyMaterial" ] }

How to contribute

Feel free to open issues or fork and submit PRs.

Verifying the PyPI package

Verifying a PyPI package is a bit complicated, but doable. Verification can be done through a chain of connected elements

  1. The cfnlambda package file found in the downloads section on PyPI

  2. The cfnlambda pgp signature also found in the downloads section on PyPI

  3. The Key ID of the person who created the signature

  4. A collection of accounts (github, twitter, etc) associated with the Key ID that illustrate that the person who signed the package is the author of the package.

You can find the package files and signatures for cfnlambda in the downloads section on PyPI. Download the package file you want to verify and the signature at the pgp link next to the package file.

Verify that the signature is a good signature by running

gpg --keyid-format long --verify cfnlambda-1.0.0.tar.gz.asc

You should get a result like this

gpg: Signature made Fri 22 May 2015 01:50:14 PM PDT
gpg:                using DSA key 0123456789ABCDEF
gpg: Can't check signature: public key not found

Now you know that the signature and the tar.gz match. Next you’ll need to verify that the person who created the signature is who you would expect. To do this look at the key ID at the end of the second line (0123456789ABCDEF in this example). That is the ID of the signatory and should be the ID of the gpg key of the author of cfnlambda. Go to keybase and type the key ID into the search bar. You should get back a single user’s profile which lists out a collection of accounts that the user has proved control of. A strong indicator that the person is the author is if you can find cfnlambda in their github account.

FAQ

Q: What causes the error inner_decorator() takes exactly 1 argument (2 given): TypeError Traceback (most recent call last): File “/var/runtime/awslambda/bootstrap.py”, line 177, in handle_event_request result = request_handler(json_input, context) TypeError: inner_decorator() takes exactly 1 argument (2 given)

A: You likely used @handler_decorator to decorate your function instead of @handler_decorator(). Because handler_decorator accepts arguments, you need to use it with parenthesis.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

cfnlambda-1.0.2.tar.gz (9.2 kB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page