AWS Lambda

Aws-SDK-Pandas

Make it easy to use pandas with AWS services.

Aws-SDK-Pandas

Create

Create the new lambda function and give the rights permissions to the role.

Configuration

TIP

Error: Runtime exited with error: signal: killed Runtime.ExitError

Provide more memory to the lambda function.

TIP

Error: Task timed out after 3.05 seconds

Increase the timeout of the lambda function.

Development

When using python with lambda, follow this pattern:

1. Python File Name: lambda_function.py

2. Invoke Function

{
    "first_name": "John",
    "last_name": "Smith"
}

3. Handler Name: lambda_handler

def hello(event):
    return message = 'Hello {} {}!'.format(event['first_name'], event['last_name']) 


def lambda_handler(event, context):
    message = hello_world(event)
    print(message)
    
    return { 
        'message' : message
    }

Python Layers

1. Create new local folder

mkdir python
cd python

2. Install desired package

TIP

Your zip file should have a folder named python with all content inside

pip3 install --platform manylinux2014_x86_64 --only-binary=:all: --upgrade -t python pyyaml

3. Create zip file

zip -r pandas_layer.zip .

4. AWS

Navigate to Lambda > Layers and create a layer.

Don`t forget to reference the compatibles runtimes, this will help you to add easily your layer to your function.