When should I use WSO2 API Microgateway instead of WSO2 API Gateway?
When you need to run a Gateway in a lock down environment or in an offline mode with zero connection to the API Management system.
When you need to run a specific subset of APIs in a controlled environment (e.g., in private jet gateway mode), when you are faced with unusual traffic patterns on an API and need to scale it.
When you need to independently scale only a specific subset of APIs based on traffic patterns.
When you are either using an internal or on-premise API Gateway deployment, where the service instances and delivery channels (consumers) reside on the same network. In this situation, you can easily reduce your latency by having your Microgateway in close proximity to your backend server.
When you are using a sidecar deployment pattern and you need to run a Gateway in sidecar mode together with an application server within a single runtime or within the same pod in Kubernetes (k8s).
When should I use WSO2 API Gateway instead of WSO2 API Microgateway?
When you need to run an API Gateway using a centralized approach so that the gateway can handle requests for multiple different APIs and backend servers.
How can I run the toolkit command from any directory location?
You need to append the full path of the /bin folder of the extracted WSO2 API Microgateway toolkit (WSO2 MGW-TK) distribution to the PATH environment variable as follows to be able to run the WSO2 API Microgateway toolkit command from any directory location. For more information, see Installation Prerequisites.
How can I run the Microgateway distribution command from any directory location?
You need to append the full path of the /bin folder of the extracted WSO2 API Microgateway distribution to the PATH environment variable as follows to be able to run the WSO2 API Microgateway distribution command from any directory location. For more information, see Installation Prerequisites.
What versions of WSO2 API-M are compatible with WSO2 API Microgateway?
Instead of using the micro-gw command, navigate to the <MGWTK_HOME>/bin directory and run the following command.
Always navigate to the <MGW_HOME>/bin directory and run the following command.
How can I reset the Microgateway CLI tool?
When you run the Microgateway CLI tool for the first time, it preserves all the information (i.e., your username, API-M base URL, TrustStore location and the TrustStore password) with the exception of your password, for later use. If you want to reset the Microgateway CLI tool in order to clear all the data that has been preserved by the Microgateway CLI tool, you can use the –reset argument as follows. Note that you can invoke the Microgateway CLI tool reset command from any directory in the system.
How can I run the Microgateway?
You can use either one of the following command formats to start WSO2 API Microgateway based on your OS.
When WSO2 Microgateway starts successfully, the following log is printed on the console.
ballerina: HTTP access log enabled
[ballerina/http] started HTTPS/WSS endpoint 0.0.0.0:9096
[ballerina/http] started HTTPS/WSS endpoint 0.0.0.0:9095
[ballerina/http] started HTTP/WS endpoint 0.0.0.0:9090
2019-05-30 18:09:32,540 INFO [wso2/gateway] - HTTPS listener is active on port 9095
2019-05-30 18:09:32,541 INFO [wso2/gateway] - HTTP listener is active on port 9090
Building a Microgateway project
What if I change the APIs or policies after I have built and deployed the project on WSO2 API Microgateway?
The gateway runtime is immutable. Therefore, if you change the APIs or policies after you have built and deployed the project on WSO2 API Microgateway, you need to rebuild the Microgateway project to capture the changes.
How can I deploy a Docker image on a Kubernetes environment?
You can use one of the following options for this purpose.
Push the Docker image to the DockerHub or to a private Docker registry.
If you use this approach and if you decide to change your Docker image name, you need to change the image name in the <MGW-project>_deployment.yaml file as well.
SCP the image to the Kubernetes nodes
Testing the Microgateway
How can I test the Microgateway?
You can invoke the API through the Microgateway using the following cURL command in order to test the WSO2 API Microgateway. By default, the Microgateway serves the HTTPS/WSS request via the 9095 port.
A signed JWT token is a self-contained access token that can be validated by the Microgateway itself without connecting to any external party Key Manager. In contrast, a typical OAuth2 token needs to be validated by connecting to the Key Manager. This is one of the key advantages of the Microgateway, if you need to run it in offline mode.
What are the ways in which I can obtain a JWT Token?
A JWT token can be obtained via one of the following methods:
From any third party secure token service. The public certificate of the token service, which is used to sign the token, should be added to the trust store of WSO2 API Microgateway. The JWT should have the claims sub, aud, exp in order for the API Microgateway to validate the JWT.
Yes, you can use an OAuth2 token instead of a JWT token. In the OAuth2 token flow you need a valid subscription for the API from WSO2 API Manager Store.
You can use a OAuth2 token instead of a JWT tokens only if the API that you are working with is also available in the WSO2 API Manager Publisher as well.
How should I configure the Key Manager when using an OAuth2 Token?
If you are using an OAuth2 access token, point the Microgateway to the Key Manager using the Key Manager details in the micro-gw.conf configuration file of the Microgateway. If you are running the Key Manager in Kubernetes, you can provide the Key Manager serverUrl as shown below. The serverUrl has to be accessible from the Microgateway pods.