CICD in AWS - GitHub to Lambda - amazon-web-services

Question

I'm trying to build a CICD pipeline that supports the very simple process laid out below. I am trying to do this all in AWS (ie, avoiding GitHub Actions), and I do not want to have to manually zip code or transfer anything.
Target process:
Git Push code to GitHub Repository.
AWS Updates code within existing Lambda function and updates the $latest alias accordingly.
Progress so far
I have been able to link AWS CodePipeline to GitHub. When code is pushed to the repository, the pipeline triggers and a compressed file that contains the contents from GitHub is added to an S3 bucket.
Long term I will likely be interested in pre- and post-deployment testing, approvals, etc etc... but for now I just want a simple setup as described above.
Challenge
I cannot fathom how then to actually update the Lamda function now I have this compressed file in S3. I've tried various Build/Deploy things from within the CodeDeloy Pipeline, but I get various errors. I'm not even entirely sure if this entire approach is the best way to go about what I want to do?!
Ask
Is this a valid approach to implementing this kind of CICD pipeline? If no, please suggest alternative and justify why you think it's better.
How do you automatically take the code from within the compressed S3 file and get it in to the Lambda function?
Thanks for your help!
Richard

Answer 1

What you could do, is include an AWS SAM (CloudFormation) template in your repository. You could then in a build step, use the build/package step of AWS SAM, which will create a packaged.yaml CloudFormation template. This template is then usable with the CloudFormation deployment actions.
This is part of a CloudFormation template that sets up such a flow, some things are omitted for brevity:
CodeBuildProject:
Type: AWS::CodeBuild::Project
Properties:
Name: codebuildproject
Description: Package and Deploy
Artifacts:
Type: CODEPIPELINE
Environment:
Type: LINUX_CONTAINER
ComputeType: BUILD_GENERAL1_SMALL
Image: aws/codebuild/amazonlinux2-x86_64-standard:3.0
EnvironmentVariables:
- Name: IAC_BUCKET
Type: PLAINTEXT
Value: !Sub iac-${AWS::Region}-${AWS::AccountId} # Bucket needed for SAM deployment
ServiceRole: !Ref CodeBuildServiceRole
Source:
Type: CODEPIPELINE
BuildSpec: |
version: 0.2
phases:
install:
runtime-versions:
python: 3.8
commands:
- 'pip install --upgrade --user aws-sam-cli'
build:
commands:
- sam build
- sam package --s3-bucket $IAC_BUCKET --output-template-file packaged.yaml
artifacts:
files:
- 'packaged.yaml'
Pipeline:
Type: AWS::CodePipeline::Pipeline
Properties:
ArtifactStore:
Location: !Sub "codepipeline-${AWS::Region}-${AWS::AccountId}"
Type: S3
Name: deployment-pipeline
RoleArn: !GetAtt PipelineExecutionRole.Arn
Stages:
- Name: Source
Actions:
- YourGithubSourceAction
- Name: Package
Actions:
- Name: SamPackage
ActionTypeId:
Category: Build
Owner: AWS
Provider: CodeBuild
Version: '1'
Configuration:
ProjectName: !Ref CodeBuildProject
InputArtifacts:
- Name: SourceZip
OutputArtifacts:
- Name: samArtifact
RunOrder: 1
- Name: Deployment
Actions:
- Name: CreateChangeSet
ActionTypeId:
Category: Deploy
Owner: AWS
Provider: CloudFormation
Version: '1'
Configuration:
ActionMode: "CHANGE_SET_REPLACE"
ChangeSetName: !Sub "${ApplicationName}-${Environment}-changeset"
Capabilities: CAPABILITY_NAMED_IAM
StackName: your-stack-name
RoleArn: !GetAtt PipelineExecutionRole.Arn
ParameterOverrides: !Sub '{ "Environment" : "${Environment}" }'
TemplatePath: 'samArtifact::packaged.yaml'
InputArtifacts:
- Name: samArtifact
RunOrder: 1
- Name: ExecuteChangeSet
ActionTypeId:
Category: Deploy
Owner: AWS
Provider: CloudFormation
Version: '1'
Configuration:
ActionMode: CHANGE_SET_EXECUTE
ChangeSetName: !Sub ${ApplicationName}-${Environment}-changeset
StackName: your-stack-name
RunOrder: 2
Be sure to have a look at AWS SAM if you're not familiar with it to see all the possibilities and how to construct your template itself.