Instead of storing api endpoints , service endpoints in gateway config , Can i use Redis database to store endpoint info .
No, this is not how Express Gateway works. The configuration must be in files as described in the documentation.
Related
We have two API servers running in HA mode i.e. same set of services are running on both VMs with same environment. We would like to use WSO2 APIM for API Security but the problem is that we have not been able to find how to use HA routing services in WSO2 APIM.
E.g.
API Server 1- http://192.168.0.2/getCustomerDetails
API Server 2- http://192.168.0.3/getCustomerDetails
API Gateway- 192.168.0.10
Once registered on API Gateway the service endpoints become-
URL1- https://192.168.0.10:8243/getCustInfo1
[edit]
URL2- https://192.168.0.10:8243/getCustInfo2
Now the question is how does WSO2 APIM decides where to route the request i.e. URL1 or URL2 for accessing the same business service? Or there is some concept like virtual ip usage in WSO2 APIM?
You don't have to create 2 APIs in API Manager for your 2 backend URLs. Create a single API and use Load Balancing or Failover Endpoints[1].
[1] https://apim.docs.wso2.com/en/latest/Learn/DesignAPI/Endpoints/high-availability-for-endpoints/
I'm new to the GCP Services, and I'm trying to make an API Gateway to proxy two services, one is run on an App Engine and the other is actually a PaaS.
Can I configure Cloud Endpoints to redirect to that PaaS, and how? and if now what service in GCP suits this case?
Edit:
An example of what I'm trying to do is if my domain is test.com then i'd like app.test.com to be redirected to my App Engine and ip.test.com would be directed for example to https://httpbin.org/ip.
I wrote an article on this for securing the endpoint. Set the value that you want in the x-google-backend.
I also provide some tips about the URL rewriting if you want.
EDIT 1
If you want to perform this routing
app.test.com -> App Engine
ip.test.com -> External service
You need to use a HTTPS Load Balancer and not API Gateway.
Then,
create a serverless NEG and configure it with App Engine
create an Internet NEG and configure it to reach your internet accessible PaaS service
Create a URL MAP with the correct routing
That's all (wait 3 - 5 minutes, the delay to dispatch your configuration around the globe.)
Yes, you can run an Extensible Service Proxy in front of your non-GCP backend service so that Cloud Endpoints can proxy requests to it.
Docs: https://cloud.google.com/endpoints/docs/openapi/running-esp-localdev
Can a HTTP endpoint be setup with AWS's Managed Streaming for Apache Kafka (MSK)? With which we can send data using HTTP POST? We can't use the AWS SDK because the client is very thin (it is our CDN actually) and it can only do simple HTTP requests like cURL/wget.
You can use the KAFKA REST Proxy within the same VPC where you are running MKS , quoting from MSK FAQ's . Refer to the Kafka REST Proxy docs for the APIs
REST Proxy: A REST proxy can be installed on an instance running within your VPC. REST proxies allow your producers and consumers to communicate to the cluster through HTTP API requests.
I have a spring boot rest project which is running successfully, and I am getting success response on http://localhost:8080/students. How can I expose this SpringBoot REST API into a private AWS API GATEWAY? I have to expose the REST API service, and I have to give the endpoints to the client, to access this service. I have no idea how to present and create endpoints in AWS API GATEWAY!
API Gateway will work as mediator between clients and your application. So in the first place your application should be exposed so AWS Gateway could make a call to it.
If you are running application on your local pc you need to expose it to Internet by setting static IP and openning corresponding ports in firewall if you have one. This may works for debug/testing purpose, but in general it is bad because you will expose your application to Internet and client services would have possibility to connect to it directly, instead of gateway.
Other and more proper on my opinion would be to run your application in AWS EC2 instance or in AWS beanstalk. In this case you can configure private endpoint for your application and make it available only through API Gateway. Here you can find more details https://aws.amazon.com/blogs/compute/introducing-amazon-api-gateway-private-endpoints/
So let's suppose you have your application up and running and available for API Gateway. In this case you will need to perform :
Create API in API Gateway
Create resources, i.e. path /user would be user resource
For each resource create a method, i.e. GET, POST, etc. Select integration type HTTP for your methods and set the endpoint it will use
Deploy API to staging
When creating an API in WSO2, the wizard has a section to pick an endpoint - current selection choices include: HTTP Endpoint, Address Endpoint, failover endpoint, Load Balance. Can additional endpoints be added, such as FTP? Is there code / config that can be added for an FTP endpoint?
WSO2 API Manager don't support FTP protocol, you can WSO2 ESB to do the same. You refer these samples here and here