Closed as not planned
Description
Is there an existing issue for this?
- I have searched the existing issues
Current Behavior
When deploying more than one (python zip) lambda to localstack, that both create boto3 SQS clients, the first one is fine but deploying the second one fails with
botocore.exceptions.ClientError: An error occurred (InternalFailure) when calling the CreateFunction operation: Unable to get handler function from lambda code: You must specify a region.
Expected Behavior
Both lambdas to deploy fine, with AWS region set by localstack
How are you starting LocalStack?
With a docker-compose file
Steps To Reproduce
My setup (simplified)
docker-compose.yml
version: '3.8'
services:
localstack:
privileged: true
image: localstack/localstack
ports:
- "4566:4566"
environment:
- LOCALSTACK_SERVICES=sqs,lambda
- LOCALSTACK_HOSTNAME=localhost
- DEFAULT_REGION=eu-west-1
Define two simple Python lambda handlers which connect to SQS with boto3:
import os
import boto3
AWS_ENDPOINT_URL = os.environ.get("AWS_ENDPOINT_URL")
sqs = boto3.resource("sqs", endpoint_url=AWS_ENDPOINT_URL)
def lambda_handler(event, context):
return "first lambda ok"import os
import boto3
AWS_ENDPOINT_URL = os.environ.get("AWS_ENDPOINT_URL")
sqs = boto3.resource("sqs", endpoint_url=AWS_ENDPOINT_URL)
def lambda_handler(event, context):
return "second lambda ok"Script to deploy two Lambdas to localstack:
import boto3
localstack_url = f"http://127.0.0.1:{EnvConfig.LOCALSTACK_PORT}"
# zip already created either manually or in other script, irrelevant for this bug
with open(_ZIP_FILENAME, "rb") as f:
zipped_code = f.read()
lambda_client = boto3.client("lambda", endpoint_url=localstack_url)
lambda_client.create_function(
FunctionName="first_lambda",
Runtime="python3.9",
Role="not_applicable",
Handler="first_lambda.lambda_handler.lambda_handler",
Code={"ZipFile": zipped_code},
PackageType="Zip",
Publish=True,
Environment={
"Variables": {
"AWS_ENDPOINT_URL": localstack_url,
}
},
)
print("first lambda deployed ok")
lambda_client.create_function(
FunctionName="second_lambda",
Runtime="python3.9",
Role="not_applicable",
Handler="second_lambda.lambda_handler.lambda_handler",
Code={"ZipFile": zipped_code},
PackageType="Zip",
Publish=True,
Environment={
"Variables": {
"AWS_ENDPOINT_URL": localstack_url,
}
},
)
# execution never gets this far
print("second lambda deployed ok")Environment
- OS: MacOS Monterey 12.2.1
- LocalStack: LatestAnything else?
As a workaround, I can set AWS_DEFAULT_REGION environment variable explicitly in all my lambdas:
Environment={
"Variables": {
"AWS_ENDPOINT_URL": localstack_url,
"AWS_DEFAULT_REGION": "eu-west-1",
}
But this shouldn't be necessary:
- localstack already has a region set
- it managed to set it for the 1st lambda
- I wouldn't need to explicitly set
AWS_DEFAULT_REGIONenv var when running the lambda in real AWS.