Build Spec and CDK Initialized

structure

auto-trading-bot
├── build-spec
│   └── docker_build_buildspec.yml
├── s3source
│   └── docker-build.zip
└── app.py

The docker build and deployment

Build Spec

First deployment in CodePipeline

./build-spec/docker_build_builspec.yml

version: 0.2

phases:
  pre_build:
    commands:
      - echo Logging into Amazon ECR
      - aws ecr get-login-password --region $AWS_DEFAULT_REGION | docker login --username AWS --password-stdin $ACCOUNT_ID.dkr.ecr.$AWS_DEFAULT_REGION.amazonaws.com
  build:
    commands:
      - echo Entered the post_build phase...
      - echo Build completed on `date`
      - docker build -t ${IMAGE_NAME} docker-build/
      - docker tag $IMAGE_NAME:latest $ECR_URI:latest
      - docker push $ECR_URI:latest

You need to add pos_build commands to update function code after the first deployment to let Lambda main function update for any changing in GitHub commits.

./build-spec/docker_build_builspec.yml

post_build:
    commands:
      - echo Update function code ...
      - aws lambda update-function-code --function-name $FUNCTION_NAME --image-uri $ECR_URI:latest --query "FunctionName"

AWS S3 Source

auto-trading-bot
├── build-spec
│   └── docker_build_buildspec.yml
├── docker-build
│   ├── app.py
│   ├── Dockerfile
│   ├── send_message.py
│   └── utils.py

Create the s3source folder contains docker-build.zip which includes all files in both folders build-spec and docker-build in zip file.

CDK intialized

Project created

Create your working folder to deploy the application in CLI terminal, activate Python environment and install required packages.

# Create working folder and CDK initialized
mkdir auto-trading-bot
cd auto-trading-bot
cdk init app --language python

# Activate Python environment
source .venv/bin/activate

# Install required packages for CDK
pip install -r requirements.txt

Configuration

CDK main function

Fill in the information prepared in Build related services for AWS.

auto-trading-bot/app.py

# Named the general namespace on your choice
namespace = <YOUR_NAMESPACE>

# IP list protection for webhook received (provided by TradingView)
webhook_ip_allowed_list = ["52.89.214.238", "34.212.75.30", "54.218.53.128", "52.32.178.7"]

# Configuration for message notification when placing the order
message_config = {
   "message_webhook_id": <YOUR_DISCORD_WEBHOOK_ID>,
   "message_webhook_token": <YOUR_DISCORD_WEBHOOK_TOKEN>
}
# Parameter format
props = {
    "SECRET_NAME": <YOUR_SECRET_NAME>,
    "REGION_NAME": <REGION_NAME>,
    "GIT_HUB_OWNER": <YOUR_GITHUB_OWNER_NAME>
    "GIT_HUB_REPO": <YOUR_REPO_NAME>,
    "GIT_HUB_BRANCH": "main",
    "WEBHOOK_SSM_NAME": "webhook_ip_list",
    "WEBHOOK_IP_ALLOWED_LIST": webhook_ip_allowed_list,
    "MESSAGE_NAME": "discord_notification",
    "MESSAGE_CONFIG": message_config,
    "namespace": namespace,
    "function_name": f"{namespace}-TradingBotLambda",
    "custom_auth_function_name": f"{namespace}-CustomAuth",
    "bucket_name": f"{namespace.lower()}-trading-bot-bucket",
    "docker_build_project_name": f"{namespace}-DockerBuild",
    "pipeline_name": f"{namespace}-Pipeline"
}

I used nested stack to split the deployment into 02 parts since we need to wait for docker image deployed on ECR at the first time deployment.

auto-trading-bot/app.py

app = cdk.App()

# Create ECR and deployment docker image first time
base_ecr = AutoTradingBotEcrStack(app, f"{props["namespace"]}-BotEcr", props)

# Create REST API Gateway, Lambda main function and Lambda Authorizer
rest_api_execution = AutoTradingBotRestApiStack(app, f"{props["namespace"]}-RestApi", base_ecr.outputs)
rest_api_execution.add_dependency(base_ecr)
app.synth()