How to modify max memory while running docker run?

[lambci]

How to modify max memory while running docker run?

[homingli]

You can use -e for setting environment variables, as referenced in Docker documentation:
https://docs.docker.com/engine/reference/run/#env-environment-variables

Example:

$ cat lambda_function.py
def lambda_handler(event, context):
    print(context.memory_limit_in_mb)
    return "It works!"

Without env var:

$ docker run --rm -v "$PWD":/var/task lambci/lambda:python3.6
START RequestId: 11038b32-74fc-40fb-9b41-120aa7065dfe Version: $LATEST
1536
END RequestId: 11038b32-74fc-40fb-9b41-120aa7065dfe
REPORT RequestId: 11038b32-74fc-40fb-9b41-120aa7065dfe Duration: 10 ms Billed Duration: 110.00000000000001 ms Memory Size: 1536 MB Max Memory Used: 18 MB
"It works!"

With env var:

$ docker run --rm -v "$PWD":/var/task -e "AWS_LAMBDA_FUNCTION_MEMORY_SIZE=128"   lambci/lambda:python3.6
START RequestId: ef3b26e2-6780-4564-a0b9-15b45b1293a4 Version: $LATEST
128
END RequestId: ef3b26e2-6780-4564-a0b9-15b45b1293a4
REPORT RequestId: ef3b26e2-6780-4564-a0b9-15b45b1293a4 Duration: 20 ms Billed Duration: 120.0 ms Memory Size: 128 MB Max Memory Used: 18 MB
"It works!"

However, I'm not sure what the behavior here is when your function exceeds the memory limit set by the env var.

[mhart]

Thanks @homingli – you're absolutely correct. To actually get docker to restrict the memory to try to simulate what would happen, you can use the -m flag with the docker run command: https://docs.docker.com/engine/reference/run/#runtime-constraints-on-resources