You can deploy your Lambda function code as a container image. AWS provides the following resources to help you build a container image for your Python function:
-
AWS base images for Lambda
- These base images are preloaded with a language runtime and other components that are required to run the image on Lambda. AWS provides a Dockerfile for each of the base images to help with building your container image.
-
Open-source runtime interface clients
- If you use a community or private enterprise base image, add a runtime interface client to the base image to make it compatible with Lambda.
Whatβs In This Document
- Create Docker image container
- Create
app.py
as lambda handler - Build and push the image to ECR
- Create Lambda container image
- Test the lambda function
π Create Docker image container
Create an image from AWS base image
FROM amazon/aws-lambda-python:3.8
COPY app.py ./
CMD ["app.handler"]
π Create app.py
as lambda handler
import sys
def handler(event, context):
return 'Hello from AWS Lambda using Python' + sys.version + '!'
π Build and push the image to ECR
aws ecr get-login-password --region ap-southeast-1 | docker login --username AWS --password-stdin 111111111111.dkr.ecr.ap-southeast-1.amazonaws.com
docker build -t pythonlambda .
docker tag pythonlambda:latest 111111111111.dkr.ecr.ap-southeast-1.amazonaws.com/pythonlambda:latest
docker push 111111111111.dkr.ecr.ap-southeast-1.amazonaws.com/pythonlambda:latest
π Create Lambda container image
π Test the lambda function
1. Invoke lambda function
- Start container
docker rm -f lambdaimagetest
docker run -d --name lambdaimagetest -it pythonlambda:latest bash
docker exec -it lambdaimagetest bash
- Invoke lambda function
β‘ $ aws lambda invoke --function-name python-lambda-image --region ap-southeast-1 outfile
{
"StatusCode": 200,
"ExecutedVersion": $LATEST"
}
β‘ $ cat outfile
"Hello from AWS Lambda using Python3.8.5 (default, Aug 5 2020, 08:22:02) \n[GCC 8.3.0]!"
2. Call lambda API from local container
- This should be performed from AWS image base
- Run container
β‘ $ docker run -d --name testapi -p 9000:8080 pythonlambda:latest
- Call API
β‘ $ curl -XPOST "http://localhost:9000/2015-03-31/functions/function/invocations" -d '{}'
"Hello from AWS Lambda using Python3.8.6 (default, Dec 16 2020, 01:05:15) \n[GCC 7.3.1 20180712 (Red Hat 7.3.1-11)]!"
- Deploy new image
- Invoke lambda function
β‘ $ aws lambda invoke --function-name python-lambda-image --region ap-southeast-1 outfile
{
"StatusCode": 200,
"ExecutedVersion": "$LATEST"
}
β‘ $ cat outfile
"Hello from AWS Lambda using Python3.8.6 (default, Dec 16 2020, 01:05:15) \n[GCC 7.3.1 20180712 (Red Hat 7.3.1-11)]!"
Mirror:
- https://github.com/vumdao/lambda-container-image
- https://vumdao.hashnode.dev/deploy-python-lambda-functions-with-container-image
Read More
- Pelican-resume with docker-compose and AWS + CDK
- Using Helm Install Botkube Integrate With Slack On EKS
- Ansible AWS EC2 Dynamic Inventory Plugin
- How To List All Enabled Regions Within An AWS account
- Using AWS KMS In AWS Lambda
- Create AWS Backup Plan
- Techniques For Writing Least Privilege IAM Policies
- EKS Persistent Storage With EFS Amazon Service
- Create k8s Cronjob To Schedule Delete Expired Files
- Amazon ECR - Lifecycle Policy Rules
- Connect Postgres Database Using Lambda Function
- Using SourceIp in ALB Listener Rule
- Amazon Simple Systems Manager (SSM)
- Invalidation AWS CDN Using Boto3
- Create AWS Lambda Function Triggered By S3 Notification Event
- CI/CD Of Invalidation AWS CDN Using Gitlab Pipeline
- Create CodeDeploy
- Gitlab Pipeline With AWS Codedeploy
- Create AWS-CDK image container
Top comments (5)
Hi, when testing with curl, where does the path "2015-03-31/functions/function/invocations" come from?
That API path is in the container image where there's process listens to, you should have a check
I am trying to follow step 1 but getting this os.environ error. Any help would be greatly appreciated.
File "/usr/local/lib/python3.8/runpy.py", line 194, in run_module_as_main
return _run_code(code, main_globals, None,
File "/usr/local/lib/python3.8/runpy.py", line 87, in _run_code
exec(code, run_globals)
File "/usr/local/lib/python3.8/site-packages/awslambdaric/main.py", line 21, in
main(sys.argv)
File "/usr/local/lib/python3.8/site-packages/awslambdaric/main.py", line 14, in main
lambda_runtime_api_addr = os.environ["AWS_LAMBDA_RUNTIME_API"]
File "/usr/local/lib/python3.8/os.py", line 675, in __getitem_
raise KeyError(key) from None
KeyError: 'AWS_LAMBDA_RUNTIME_API'
you should use aws-lambda-rie (emulator) instead of aws-lambda-ric - docs.aws.amazon.com/lambda/latest/.... I believe that environment variable is set by AWS, so if it's not present that script will execute the emulator. This is an example
entry_script.sh
Dockerfile
Hi Jason,
I guess you're trying to build lambda image from alternative base image but I recommend to use image base from AWS such as
public.ecr.aws/lambda/python:3.8
oramazon/aws-lambda-python:3.8
For your issue, you're missing setting
ENV AWS_LAMBDA_RUNTIME_API=
, this variable env must be defined but you will still get following error after that