I will answer those points that are not based on opinions:
- An elastic beanstalk gets a subdomain when creating an application. This subdomain should be used by the Gateway API with the integration type: AWS service, in the action configuration - Am I right?
No, AWS integration is only applicable if you are actually invoking Elastic Beanstalk . You will call your own beanstalk instance to use HTTP integration.
- What will be one microservice? Is Elastic Beanstalk a specific scalable microservice service?
It is up to you, but as mentioned in the comments, many clients prefer to do this through Lambda functions rather than beanstalk applications. Using Lambda has the advantage that you do not need to control the scaling of your beanstalk application.
- Test conditions: what structure should be used in a test environment (or setting).? I am thinking of creating a separate VPC with another Elastic bobbin and other Amazon services.
Just note that the Gateway API currently cannot communicate with your VPC resources. Any beanstalk instance must be publicly available.
- Test environment and API gateway: how do I configure an API gateway? It should allow clients to access microservices in a test environment if the request has a specific subdomain, for example: test.mydomain.com/hello_world/say_hello. I am not sure how to use the Gateway API in CI / CD to make it quick and easy, without manually copying some configuration from the testing phase to the production phase. (I do not expect any complicated solution, only some tips about which components, parts and concepts I can use for it. I will find more details on my own).
You should take a look at the stage variables . This will allow you to use the basic configuration with the differences between dev / test / prod stored in these variables.