Lambda layers provide a convenient way to package libraries and other dependencies that you can use with your Lambda functions. Using layers reduces the size of uploaded deployment archives and makes it faster to deploy your code. Layers also promote code sharing and separation of responsibilities so that you can iterate faster on writing business logic.
A layer is a .zip file archive that can contain additional code or data. A layer can contain libraries, a custom runtime, data, or configuration files. Layers promote code sharing and separation of responsibilities so that you can iterate faster on writing business logic.
Created this action to save on all the extra code that is required in a GitHub Workflow to create and publish a layer. This will do all the undifferentiated heavy lifting for you.
Using this action requires you to add a block of code to your Github workflow.
| Name | Type | Required | Description |
|---|---|---|---|
layer-name |
string | Yes | SSM Parameter Name |
layer-directory |
string | Yes | Working directory in repository where requirements.txt file exists |
runtime |
string | Yes | AWS Lambda Runtime. Allowed Values: python, node |
bucketname |
string | Yes | AWS S3 Bucket Name where layer will be uploaded |
prefix-folder-path |
string | No | (Optional) An optional prefix that will be used for a folder path inside the S3 bucket |
principal |
string | Yes | An AWS Account ID to grant layer usage permissions to |
organization-id |
string | No | An AWS Organization ID to grant layer usage permissions to. Principal must be "*" if this is set. |
As as emample, suppose you want to create an AWS Lambda Layer for your lambda function. You can see in the example below some sameple values and they can be shown in an example below for directory layout.
- name: Build and Create Layer
uses: stephenbawks/aws-lambda-layer@v0.2.0
with:
layer-name: my-awesome-layer-name
layer-directory: layer
runtime: python
bucketname: my-s3-bucket-name-goes-here
prefix-folder-path: beta
principal: "123456789101" - name: Build and Create Layer
uses: stephenbawks/aws-lambda-layer@v0.2.0
with:
layer-name: my-awesome-layer-name
layer-directory: layer
runtime: python
bucketname: my-s3-bucket-name-goes-here
prefix-folder-path: beta
principal: "*" - name: Build and Create Layer
uses: stephenbawks/aws-lambda-layer@v0.2.0
with:
layer-name: my-awesome-layer-name
layer-directory: layer
runtime: python
bucketname: my-s3-bucket-name-goes-here
prefix-folder-path: beta
principal: "*"
organization-id: "o-1234567890"The layer-name can be whatever you want to call your layer. This is unique to you of course and the layer name can contain only letters, numbers, hyphens, and underscores.
The layer-directory is more opinionated in its layout. In this example, in the root of your repository you would have a directory that is called layer and under that directory it would contain your requirements.txt file with your dependencies you want to bundle up in the layer. The action is also using this same directory as a working directory to download your requirements and then created the zip file that is used for the layer. We want to keep it segmented so that its not pulling extra files/folders that accidently get added to your layer.
The runtime only accepts node and python at this time. More to be added later.
The bucketname argument is the name of your S3 bucket where you want the layer to be uploaded to.
The variable prefix-folder-path is there to add an extra folder in the folder path where your layer is stored in your S3 bucket. This variable is optional and not required but could be useful if you want to have different version of your layer stored in different locations.
The default path schema for the layer looks like the following:
s3://<bucketname>/<prefix-folder-path>/<layer-name>-<DATE>-<RANDOM_NUMBER>.zip
And as an example would result in a path that looks like the following:
s3://my-s3-bucket-name-goes-here/beta/my-awesome-layer-name-29-12-2021-34522.zip
The principal variable is the account number you want to share the layer with. If you put "*" here that will share it to the world.
The organization-id grants permission to all accounts in the specified organization. The principal needs to be set to *.
The resulting layer name will look like the following.
arn:aws:lambda:<AWS_REGION>:<AWS_ACCOUNT_ID>:layer:<layer-name>-<mm-dd-yyy>:<layer_version>
For each Lambda runtime, the PATH variable includes specific folders in the /opt directory. If you define the same folder structure in your layer .zip file archive, your function code can access the layer content without the need to specify the path.
The following table lists the folder paths that each runtime supports and the directory the action will be creating and installing your requirements under.
| Runtime | Path |
|---|---|
node |
nodejs/node_modules |
python |
python |
There is also some example requirements file as well as a Github Workflow file you can check out here to get yourself started quickly.
| Example | Link |
|---|---|
| Github Workflow | Link |
| Node Requirements | Link |
| Python Requirements | Link |