🐾Lambda development - Container function🐾

❓All Lambda layers should take no more than 250 Mb in total and it can be a problem with heavy ML libraries. With Container functions get 10 GB for your libraries and other artefacts.

Steps to create Docker image for Lambda and push it to ECR:

1️⃣ Create Dockerfile and specify AWS base image to start with. You can use your own base image, but you need to install the Python runtime interface client.

FROM public.ecr.aws/lambda/python:3.9

2️⃣ Install dependencies from requirements.txt file. If you don’t want to create requirements.txt file, you can install dependencies with pip3 command.

COPY requirements.txt . 
RUN pip3 install -r requirements.txt --target "${LAMBDA_TASK_ROOT}"

3️⃣ Copy you Python code and all other dependencies to container

COPY app.py ${LAMBDA_TASK_ROOT}
COPY artifact.json ${LAMBDA_TASK_ROOT}

4️⃣ Set the CMD, so your container can find and run your handler function (your app.py code should have the same structure as your usual Lambda function code)

CMD [ "app.handler" ]

5️⃣ Login to the ECR using the following command (otherwise you won’t be able to push the image)

docker login -u AWS -p ${aws ecr get-login-password --region <YOUR_REGION>} <YOUR_ACCOUNT>.dkr.ecr.<>YOUR_REGION.amazonaws.com

6️⃣ Run command to build your docker image and push it to ECR

docker build -t docker push <AWS_ECR_REPOSITORY_URL>:latest .
docker push <AWS_ECR_REPOSITORY_URL>:latest

---

If you like this post, you can share APAWS newsletter with friends: