iot_rekognition_demo

Demo of Greengrass Iot PiCam and Rekognition

Requirements

AWS CLI already configured with at least PowerUser permission
Python 3 installed
Pipenv installed
- pip install pipenv
Docker installed
SAM Local installed

Provided that you have requirements above installed, proceed by installing the application dependencies and development dependencies:

pipenv install
pipenv install -d

Testing

Pytest is used to discover tests created under tests folder - Here's how you can run tests our initial unit tests:

pipenv run python -m pytest tests/ -v

Tip: Commands passed to pipenv run will be executed in the Virtual environment created for our project.

Packaging

AWS Lambda Python runtime requires a flat folder with all dependencies including the application. To facilitate this process, the pre-made SAM template expects this structure to be under <src>/build/:

...
    FirstFunction:
        Type: AWS::Serverless::Function
        Properties:
            CodeUri: first_function/build/
            ...

With that in mind, we will:

Generate a hashed requirements.txt out of our Pipfile dep file
Install all dependencies directly to build sub-folder
Copy our function (app.py) into build sub-folder

# Create a hashed pip requirements.txt file only with our app dependencies (no dev deps)
pipenv lock -r > requirements.txt
pip install -r requirements.txt -t first_function/build/
cp -R first_function/app.py first_function/build/

Local development

Given that you followed Packaging instructions then run one of the following options to invoke your function locally:

Invoking function locally without API Gateway

echo '{"lambda": "payload"}' | sam local invoke FirstFunction

Invoking function locally through local API Gateway

sam local start-api

If the previous command run successfully you should now be able to hit the following local endpoint to invoke your function http://localhost:3000/first/REPLACE-ME-WITH-ANYTHING.

Deployment

First and foremost, we need a S3 bucket where we can upload our Lambda functions packaged as ZIP before we deploy anything - If you don't have a S3 bucket to store code artifacts then this is a good time to create one:

aws s3 mb s3://BUCKET_NAME

Provided you have a S3 bucket created, run the following command to package our Lambda function to S3:

aws cloudformation package \
    --template-file template.yaml \
    --output-template-file packaged.yaml \
    --s3-bucket REPLACE_THIS_WITH_YOUR_S3_BUCKET_NAME

Next, the following command will create a Cloudformation Stack and deploy your SAM resources.

aws cloudformation deploy \
    --template-file packaged.yaml \
    --stack-name iot_mirror \
    --capabilities CAPABILITY_IAM

See Serverless Application Model (SAM) HOWTO Guide for more details in how to get started.

After deployment is complete you can run the following command to retrieve the API Gateway Endpoint URL:

aws cloudformation describe-stacks \
    --stack-name iot_mirror \
    --query 'Stacks[].Outputs'

Appendix

AWS CLI commands

AWS CLI commands to package, deploy and describe outputs defined within the cloudformation stack:

aws cloudformation package \
    --template-file template.yaml \
    --output-template-file packaged.yaml \
    --s3-bucket REPLACE_THIS_WITH_YOUR_S3_BUCKET_NAME

aws cloudformation deploy \
    --template-file packaged.yaml \
    --stack-name iot_mirror \
    --capabilities CAPABILITY_IAM \
    --parameter-overrides MyParameterSample=MySampleValue

aws cloudformation describe-stacks \
    --stack-name iot_mirror --query 'Stacks[].Outputs'

boul / iot-rekognition-demo