I was able to setup localstack (https://github.com/atlassian/localstack) and also create lambda function in it (using create-function ... command). However, I couldnt find a way
Looks like there is an open issue related to setting up API Gateway with localstack on GitHub:
https://github.com/localstack/localstack/issues/129
You could try following the steps provided in the answer there.
Copied from the GitHub issue:
"""
One option would be to use the serverless framework (https://github.com/serverless/serverless). Otherwise, you can call the LocalStack services directly (via the CLI or an SDK) to create an API Gateway resource+method+integration, and connect them to your Lambda function.
Here are a few pointers that might be helpful: https://ig.nore.me/2016/03/setting-up-lambda-and-a-gateway-through-the-cli/ (the "Creating a role" part can be skipped) https://github.com/atlassian/localstack/issues/101 https://github.com/temyers/serverless-localstack """
Walkthrough for creating a NodeJS Lambda together with API Gateway per CLI:
First we create a simple NodeJS Lambda:
const apiTestHandler = (payload, context, callback) => {
console.log(`Function apiTestHandler called with payload ${JSON.stringify(payload)}`);
callback(null, {
statusCode: 201,
body: JSON.stringify({
somethingId: payload.pathParameters.somethingId
}),
headers: {
"X-Click-Header": "abc"
}
});
}
module.exports = {
apiTestHandler,
}
Put that into a zip File called apiTestHandler.zip and upload it to localstack:
aws lambda create-function \
--region us-east-1 \
--function-name api-test-handler \
--runtime nodejs6.10 \
--handler index.apiTestHandler \
--memory-size 128 \
--zip-file fileb://apiTestHandler.zip \
--role arn:aws:iam::123456:role/role-name --endpoint-url=http://localhost:4574
Now we can create our Rest-Api:
aws apigateway create-rest-api --region us-east-1 --name 'API Test' --endpoint-url=http://localhost:4567
This gives the following response:
{
"name": "API Test",
"id": "487109A-Z548",
"createdDate": 1518081479
}
With the ID we got here, we can ask for its parent-ID:
aws apigateway get-resources --region us-east-1 --rest-api-id 487109A-Z548 --endpoint-url=http://localhost:4567
Response:
{
"items": [
{
"path": "/",
"id": "0270A-Z23550",
"resourceMethods": {
"GET": {}
}
}
]
}
Now we have everything to create our resource together with its path:
aws apigateway create-resource \
--region us-east-1 \
--rest-api-id 487109A-Z548 \
--parent-id 0270A-Z23550 \
--path-part "{somethingId}" --endpoint-url=http://localhost:4567
Response:
{
"resourceMethods": {
"GET": {}
},
"pathPart": "{somethingId}",
"parentId": "0270A-Z23550",
"path": "/{somethingId}",
"id": "0662807180"
}
The ID we got here is needed to create our linked GET Method:
aws apigateway put-method \
--region us-east-1 \
--rest-api-id 487109A-Z548 \
--resource-id 0662807180 \
--http-method GET \
--request-parameters "method.request.path.somethingId=true" \
--authorization-type "NONE" \
--endpoint-url=http://localhost:4567
We are almost there - one of the last things to do is to create our integration with the already uploaded lambda:
aws apigateway put-integration \
--region us-east-1 \
--rest-api-id 487109A-Z548 \
--resource-id 0662807180 \
--http-method GET \
--type AWS_PROXY \
--integration-http-method POST \
--uri arn:aws:apigateway:us-east-1:lambda:path/2015-03-31/functions/arn:aws:lambda:us-east-1:000000000000:function:api-test-handler/invocations \
--passthrough-behavior WHEN_NO_MATCH \
--endpoint-url=http://localhost:4567
Last but not least: Deploy our API to our desired stage:
aws apigateway create-deployment \
--region us-east-1 \
--rest-api-id 487109A-Z548 \
--stage-name test \
--endpoint-url=http://localhost:4567
Now we can test it:
curl http://localhost:4567/restapis/487109A-Z548/test/_user_request_/HowMuchIsTheFish
Response:
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 34 100 34 0 0 9 0 0:00:03 0:00:03 --:--:-- 9
{"somethingId":"HowMuchIsTheFish"}
I hope this helps.
Hint 1: For easier use I recommend to install AWSCLI Local ( https://github.com/localstack/awscli-local ) - with this tool you can use the command "awslocal" and don't have to type "--endpoint-url= ..." for each command
Walkthrough for using Serverless Framework and Localstack:
You can also use the Serverless Framework (https://serverless.com/).
First install it via npm:
npm install serverless -g
Now you can create a sample application based on a nodejs-aws template:
serverless create --template aws-nodejs
In order to have an HTTP endpoint, you have to edit the serverless.yml
and add the corresponding event :
functions:
hello:
handler: handler.hello
events:
- http:
path: ping
method: get
In order to run this against your localstack installation you have to use the serverless-localstack plugin ( https://github.com/temyers/serverless-localstack):
npm install serverless-localstack
Now you have to edit your serverless.yml
again, add the plugin and adjust your endpoints. In my case localstack is running inside the Docker toolbox, so it's IP is 192.168.99.100
- you may have to change this to localhost
, depending on your use:
plugins:
- serverless-localstack
custom:
localstack:
debug: true
stages:
- local
- dev
host: http://192.168.99.100
endpoints:
S3: http://192.168.99.100:4572
DynamoDB: http://192.168.99.100:4570
CloudFormation: http://192.168.99.100:4581
Elasticsearch: http://192.168.99.100:4571
ES: http://192.168.99.100:4578
SNS: http://192.168.99.100:4575
SQS: http://192.168.99.100:4576
Lambda: http://192.168.99.100:4574
Kinesis: http://192.168.99.100:4568
Now you can try to deploy it:
serverless deploy --verbose --stage local
This will create an S3 bucket, upload your lambda and create a cloudformation stack. However, the process will fail due to some inconsistencies of localstack when compared against AWS. Don't be dismayed though, the created cloudformation template works fine and you just need an additional request and you are done:
awslocal cloudformation update-stack --template-body file://.serverless/cloudformation-template-update-stack.json --stack-name aws-nodejs-local
Now your lambda is deployed and can be tested:
curl http://192.168.99.100:4567/restapis/75A-Z278430A-Z/local/_user_request_/ping
Response:
% Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed
100 364 100 364 0 0 111 0 0:00:03 0:00:03 --:--:-- 111
{"message":"Go Serverless v1.0! Your function executed successfully!","input":{"body":null,"headers":{"host":"192.168.99.100:4567","accept":"*/*","user-agent":"curl/7.49.1"},"resource":"/restapis/75A-Z278430A-Z/local/_user_request_/ping","queryStringParameters":{},"httpMethod":"GET","stageVariables":{},"path":"/ping","pathParameters":{},"isBase64Encoded":false}}
Hope this helps.