Lambda layers for shared code in AWS - Commands & Configuration

Introduction

When you have code or libraries that many AWS Lambda functions need, copying them into each function is slow and wasteful. Lambda layers let you put shared code in one place and use it in many functions easily.

When multiple Lambda functions use the same helper code or libraries.

When you want to update shared code without changing each Lambda function separately.

When you want to keep your Lambda function packages small and focused.

When you want to share common runtime dependencies across functions.

When you want to organize your code better by separating shared parts.

Config File - layer.zip

layer.zip

This is a zip file containing your shared code or libraries structured properly for Lambda layers. For example, for Python, it should have a folder named 'python' with your modules inside. This zip file is uploaded to AWS Lambda as a layer.

The zip file contains the shared code organized in a folder structure that Lambda expects. For Python, the 'python' folder is where your .py files or packages go. This lets Lambda add the layer's code to the function's runtime environment automatically.

Commands

This command uploads your zip file as a new Lambda layer version named 'my-shared-layer'. It tells Lambda this layer works with Python 3.9 runtime.

Terminal

aws lambda publish-layer-version --layer-name my-shared-layer --description "Shared utilities" --zip-file fileb://layer.zip --compatible-runtimes python3.9

Expected Output

{

    "LayerVersionArn": "arn:aws:lambda:us-east-1:123456789012:layer:my-shared-layer:1",

    "Version": 1,

    "Description": "Shared utilities",

    "CreatedDate": "2024-06-01T12:00:00.000+0000",

    "CompatibleRuntimes": [

        "python3.9"

    ]

}

→

--layer-name - Names your Lambda layer.

→

--zip-file - Specifies the zip file with your shared code.

→

--compatible-runtimes - Defines which Lambda runtimes can use this layer.

This command attaches the published layer version to your existing Lambda function named 'my-function'. This makes the shared code available to that function.

Terminal

aws lambda update-function-configuration --function-name my-function --layers arn:aws:lambda:us-east-1:123456789012:layer:my-shared-layer:1

Expected Output

{

    "FunctionName": "my-function",

    "Layers": [

        {

            "Arn": "arn:aws:lambda:us-east-1:123456789012:layer:my-shared-layer:1",

            "CodeSize": 12345

        }

    ]

}

→

--function-name - Specifies which Lambda function to update.

→

--layers - Lists the layer ARNs to attach to the function.

This command checks the current configuration of your Lambda function to confirm the layer is attached.

Terminal

aws lambda get-function-configuration --function-name my-function

Expected Output

{

    "FunctionName": "my-function",

    "Layers": [

        {

            "Arn": "arn:aws:lambda:us-east-1:123456789012:layer:my-shared-layer:1",

            "CodeSize": 12345

        }

    ],

    "Runtime": "python3.9"

}

→

--function-name - Specifies which Lambda function to inspect.

Key Concept

If you remember nothing else from this pattern, remember: Lambda layers let you share code once and use it in many functions without copying it each time.

Common Mistakes

Uploading the zip file without the correct folder structure inside (like missing the 'python' folder for Python code).

Lambda won't find your shared code because it expects a specific folder layout inside the layer zip.

Make sure your zip file has the correct folder structure, for example, a 'python' folder containing your Python modules.

Not specifying compatible runtimes when publishing the layer.

Your layer might not be usable by your Lambda function if the runtime is not listed as compatible.

Always include the --compatible-runtimes flag with the runtimes your layer supports.

Forgetting to update the Lambda function configuration to attach the new layer version.

Your function won't use the shared code until the layer is attached.

Run the update-function-configuration command with the layer ARN to attach the layer.

Summary

Create a zip file with your shared code organized in the correct folder structure.

Publish the zip as a Lambda layer with compatible runtimes specified.

Attach the published layer version to your Lambda function using its ARN.

Verify the function configuration to confirm the layer is attached.

Practice

1. What is the main purpose of using Lambda layers in AWS Lambda functions?

easy

A. To share common code and libraries across multiple Lambda functions

B. To increase the memory size of a Lambda function

C. To schedule Lambda functions to run at specific times

D. To monitor Lambda function execution logs

Solution

1. Step 1: Understand Lambda layers purpose

   Lambda layers are designed to hold shared code or libraries that multiple Lambda functions can use.

2. Step 2: Compare options with purpose

   Options A, C, and D describe unrelated Lambda features like memory, scheduling, and logging.

3. Final Answer:

   To share common code and libraries across multiple Lambda functions -> Option A

4. Quick Check:

   Lambda layers = shared code [OK]

Hint: Layers share code across functions, not manage memory or logs [OK]

Common Mistakes:

• Confusing layers with memory settings
• Thinking layers schedule functions
• Mixing layers with logging features

2. Which of the following is the correct way to add a Lambda layer to a function using AWS CLI?

easy

A. aws lambda update-function-configuration --function-name MyFunction --layers arn:aws:lambda:region:account-id:layer:MyLayer:1

B. aws lambda create-layer-version --function-name MyFunction --layer-name MyLayer

C. aws lambda add-layer-version --function-name MyFunction --layer-name MyLayer

D. aws lambda attach-layer --function-name MyFunction --layer-name MyLayer

Solution

1. Step 1: Identify correct CLI command to add layer

   The correct command to add a layer to a Lambda function is update-function-configuration with the --layers parameter specifying the layer ARN.

2. Step 2: Review options for syntax correctness

   aws lambda update-function-configuration --function-name MyFunction --layers arn:aws:lambda:region:account-id:layer:MyLayer:1 uses the correct command and syntax. Options A, B, and D use invalid or non-existent commands.

3. Final Answer:

   aws lambda update-function-configuration --function-name MyFunction --layers arn:aws:lambda:region:account-id:layer:MyLayer:1 -> Option A

4. Quick Check:

   Update function config with layers ARN = aws lambda update-function-configuration --function-name MyFunction --layers arn:aws:lambda:region:account-id:layer:MyLayer:1 [OK]

Hint: Use update-function-configuration with --layers ARN [OK]

Common Mistakes:

• Using create-layer-version to attach layers
• Trying non-existent commands like add-layer-version
• Not specifying full ARN for the layer

3. Given this Python Lambda function code snippet using a layer for shared utilities:

import shared_utils

def lambda_handler(event, context):
    return shared_utils.process(event)

What will happen if the Lambda function is invoked but the layer is not attached?

medium

A. The function will timeout waiting for the layer

B. The function will run successfully without errors

C. The function will return None silently

D. The function will raise a ModuleNotFoundError for 'shared_utils'

Solution

1. Step 1: Understand import behavior without layer

   If the layer providing shared_utils is missing, Python cannot find the module and raises an error.

2. Step 2: Identify the error type

   Python raises ModuleNotFoundError when an import fails.

3. Final Answer:

   The function will raise a ModuleNotFoundError for 'shared_utils' -> Option D

4. Quick Check:

   Missing layer = ModuleNotFoundError [OK]

Hint: Missing layer means missing module, causing import error [OK]

Common Mistakes:

• Assuming function runs without the layer
• Expecting silent failure or None return
• Thinking function times out due to missing layer

4. You updated your Lambda layer code and published a new version. After updating your function configuration to use the new layer version, the function still uses old code. What is the most likely cause?

medium

A. The function needs a restart to pick up the new layer version

B. The function configuration was not updated to the new layer version ARN

C. The new layer version was not published correctly

D. Lambda layers automatically update all functions, so this is impossible

Solution

1. Step 1: Understand layer versioning

   Each layer version has a unique ARN. Functions must explicitly reference the new version ARN to use updated code.

2. Step 2: Identify common mistake

   If the function still uses old code, it likely still references the old layer version ARN.

3. Final Answer:

   The function configuration was not updated to the new layer version ARN -> Option B

4. Quick Check:

   Update function config with new layer ARN to use new code [OK]

Hint: Update function config with new layer ARN after publishing [OK]

Common Mistakes:

• Assuming layers auto-update all functions
• Not publishing new layer version properly
• Thinking function restart is needed

5. You have two Lambda functions, both using a shared layer with version 3. You want to update the shared code without affecting the functions immediately. What is the best practice to update the shared code safely?

hard

A. Delete version 3 and create a new layer with the same version number

B. Overwrite version 3 of the layer with new code so both functions get updates automatically

C. Publish a new layer version 4 with updated code, then update each function to use version 4 when ready

D. Update the functions' code directly instead of using layers

Solution

1. Step 1: Understand layer version immutability

   Lambda layer versions are immutable; you cannot overwrite an existing version.

2. Step 2: Apply best practice for safe updates

   Publish a new version with updated code and update functions individually to control rollout.

3. Final Answer:

   Publish a new layer version 4 with updated code, then update each function to use version 4 when ready -> Option C

4. Quick Check:

   New layer version + update functions individually = safe update [OK]

Hint: Publish new layer version; update functions selectively [OK]

Common Mistakes:

• Trying to overwrite existing layer version
• Deleting and recreating layers with same version
• Updating function code instead of using layers