Bug: sam local start-lambda not work in SDK BOTO3

[b-y-f]

Description:

So strange after run sam local start-lambda, I can invoke lambda with cli aws lambda invoke --function-name "myFunctionName" --endpoint-url "http://127.0.0.1:3001" --no-verify-ssl --cli-binary-format raw-in-base64-out --payload '{"xx":"xx"}' out.txt, It gives 200 code.

But when I use boto3 with code like below

lambda_client = boto3.client(
    "lambda",
    endpoint_url="http://127.0.0.1:3001"
    )
res = lambda_client.invoke(
    FunctionName="myFunctionName", Payload=json.dumps({"xxx": "xxx"})
)

Error like below raised.

PLUS: after that, I uploaded the lambda I want to test to cloud and used boto.client('lambda') and invoked without any problem.

Steps to reproduce:

• No 3rd party, just vscode + aws toolkits
• everything follows instruction from aws doc.
• Dev under mini conda env

Observed result:

START RequestId: 2c6dcc30-1b2e-4af6-8b1f-26658d0eb0f1 Version: $LATEST
[ERROR] EndpointConnectionError: Could not connect to the endpoint URL: "http://127.0.0.1:3001/2015-03-31/functions/fetchGeoDataB    raise EndpointConnectionError(endpoint_url=request.url, error=e)ponset_exception
{"errorMessage": "Could not connect to the endpoint URL: \"http://127.0.0.1:3001/2015-03-31/functions/fetchGeoDataByICP/invocations\"", "errorType": "EndpointConnectionError", "requestId": "2c6dcc30-1b2e-4af6-8b1f-26658d0eb0f1", "stackTrace": ["  File \"/var/task/app.py\", line 15, in lambda_handler\n    res = lambda_client.invoke(\n", "  File \"/var/runtime/botocore/client.py\", line 391, in _api_call\n    return self._make_api_call(operation_name, kwargs)\n", "  File \"/var/runtime/botocore/client.py\", line 705, in _make_api_call\n    http, parsed_response = self._make_request(\n", "  File \"/var/runtime/botocore/client.py\", line 725, in _make_request\n    return self._endpoint.make_request(operation_model, request_dict)\n", "  File \"/var/runtime/botocore/endpoint.py\", line 104, in make_request\n    return self._send_request(request_dict, operation_model)\n", "  File \"/var/runtime/botocore/endpoint.py\", line 138, in _send_request\n    while self._needs_retry(attempts, operation_model, request_dict,\n"END RequestId: 2c6dcc30-1b2e-4af6-8b1f-26658d0eb0f1
REPORT RequestId: 2c6dcc30-1b2e-4af6-8b1f-26658d0eb0f1  Init Duration: 0.14 ms  Duration: 9938.23 ms    Billed Duration: 9939 msMemory Size: 128 MB      Max Memory Used: 128 MB
, "  File \"/var/runtime/botocore/endpoint.py\", line 254, in _needs_retry\n    responses = self._event_emitter.emit(\n", "  File \"/var/runtime/botocore/hooks.py\", line 357, in emit\n    return self._emitter.emit(aliased_event_name, **kwargs)\n", "  File \"/var/runtime/botocore/hooks.py\", line 228, in emit\n    return self._emit(event_name, kwargs)\n", "  File \"/var/runtime/botocore/hooks.py\", line 211, in _emit\n    response = handler(**kwargs)\n", "  File \"/var/runtime/botocore/retryhandler.py\", line 183, in __call__\n    if self._checker(attempts, response, caught_exception):\n", "  File \"/var/runtime/botocore/retryhandler.py\", line 250, in __call__\n    should_retry = self._should_retry(attempt_number, response,\n", "  File \"/var/runtime/botocore/retryhandler.py\", line 277, in _should_retry\n    return self._checker(attempt_number, response, caught_exception)\n", "  File \"/var/runtime/botocore/retryhandler.py\", line 316, in __call__\n    checker_response = checker(attempt_number, response,\n", "  File \"/var/runtime/botocore/retryhandler.py\", line 222, in __call__\n    return self._check_caught_exception(\n", "  File \"/var/runtime/botocore/retryhandler.py\", line 359, in _check_caught_exception\n    raise caught_exception\n", "  File \"/var/runtime/botocore/endpoint.py\", line 201, in _do_get_response\n    http_response = self._send(request)\n", "  File \"/var/runtime/botocore/endpoint.py\", line 270, in _send\n    return self.http_session.send(request)\n", "  File \"/var/runtime/botocore/httpsession.py\", line 438, in send\n    raise EndpointConnectionError(endpoint_url=request.url, error=e)\n"]}%                                 

Expected result:

It should run my lambda and return something not EndpointConnectionError

Additional environment details (Ex: Windows, Mac, Amazon Linux etc)

1. OS: MAC M1
2. If using SAM CLI, sam --version: version 1.52.0
3. AWS region:ap-southeast-2

Add --debug flag to any SAM CLI commands you are running

[jfuss]

Moving to correct repo.

[mildaniel]

Hey @b-y-f, I tried reproducing this issue on a non-M1 Mac and wasn't able to. I will try again on an M1 machine to see if the behaviour is different.

[b-y-f]

Hey @b-y-f, I tried reproducing this issue on a non-M1 Mac and wasn't able to. I will try again on an M1 machine to see if the behaviour is different.

Thank you

[jfuss]

@b-y-f This has to do with docker networking. http://127.0.0.1:3001 doesn't resolve to your machine's localhost but the docker machine's localhost. Since sam local start-lambda is running on your machine, this is why you get a response using the CLI but fails within the container. You can follow #510 (comment) for details.

[b-y-f]

Thankyou