Error occured. (arm64, docker)

[sdg9670]

Env: ubuntu, arm64, docker(node 18 image)

// test.js
exports.handler = async (event) => {
    const response = {
        statusCode: 200,
        body: JSON.stringify('Hello from Lambda!'),
    }
    return response
}

npm install aws-lambda-ric
npx aws-lambda-ric test.handler

I got this error.

Executing 'test.handler' in function directory '/workspaces/test'
terminate called after throwing an instance of 'std::logic_error'
  what():  basic_string: construction from null is not valid
Aborted (core dumped)

how can I solve this problem?

[amitizle]

We're having the same issue, @sdg9670 did you manage to resolve it?

[amitizle]

We resolved it by using [email protected] and not the latest version.

[sdg9670]

@amitizle I gave up to use arm64 cpu. lol...

[amunhoz]

I'm having the same problem in amd64 with latest version, any clue?

[amitizle]

@amunhoz see my comment in here - try and use version 2.1.0

[schuettc]

We resolved it by using [email protected] and not the latest version.

This is working for me. But quite difficult to figure out. Would like an update on when ^3 will work with arm64

[carbontwelve]

I get the same error while following the official guide on using an alternative base image with the runtime interface client: https://docs.aws.amazon.com/lambda/latest/dg/nodejs-image.html#nodejs-image-clients

This is on x86, node20.

[sin-ack]

The cause of this is this bit of code:

aws-lambda-nodejs-runtime-interface-client/src/rapid-client.cc

Line 24 in db6a30d

static const std::string ENDPOINT(getenv("AWS_LAMBDA_RUNTIME_API"));

This uses getenv in a global initializer to get an environment variable during the load of the RIC. If the environment variable is not set, the STL will throw an exception, and Node.js has no way of recovering from that (since it happens at module load time). The correct fix is to check for the environment variable in Init and to just throw a JavaScript-level exception explaining that the environment variable must be set. You should be able to bypass this by setting the AWS_LAMBDA_RUNTIME_API environment variable for now.