theangrydev / TrafficParrot

This repo is to have Trafffic parrot mock server code

MOCK service setup using Traffic Parrot

In this project, we will be setting up Traffic Parrot to simulate REST and GRPC api mock servers

Traffic parrot is an API mocking and service virtualization tool. It simulates APIs and services so that you can test your microservice without having to worry about test data set up or environment availability.

To experience Traffic Parrot locally follow these instructions

Deploy deploy Debug Resolver mock service to k3d

1. Pre-requisites

   1. install docker (Create docker login, if you don't have one)
      install kubectl
      install k3d
      	wget -q -O - https://raw.githubusercontent.com/rancher/k3d/master/install.sh | bash
      	This step will install k3d under root and also default namespace created in root.
      	However to view the work pods in user's name space please follow the below steps
      	k3d delete       
      

2. k3d instructions to setup new cluster and name space:

   1. k3d create \
          --publish 8080:80 \
          --publish 8443:443 \
          --workers 3 \
          --server-arg --no-deploy=traefik
      
      

   export KUBECONFIG="$(k3d get-kubeconfig --name='k3s-default')"

   #Verify k3d setup
   kubectl get pods
   kubectl config get-clusters
   kubectl get nodes 
   ```
   

3. From step-3, we start on mock service setup. Please follow the below check list to get mock service running.

• Unzip the TrafficParrot.zip file

• For GRPC apis, the spec will be defined in proto files. So, to create mock service, please collect the list of .proto files from dev team required for your service

• Typically, ALL the GRPC api based services will have main proto file. This will be main source for all the GRPC apis.

• What could be the challenge here? Normally in the main proto file, there might other dependent import of proto files. like below import "github.com/lyft/protoc-gen-validate/validate/validate.proto"; import "google/api/annotations.proto"; import "protoc-gen-swagger/options/annotations.proto"; So, to create mock service setup, we need to have all the dependent proto files with same folder structure.

• Once you have all the proto files, copy them to proto folder of traffic parrot

• Run the ./start.sh file in traffic parrot folder.

• Locad the hostname:8080 url in the browser, and your Traffic parrot is ready.

• Click on GRPC and select Add/edit option in the UI

• Now the time to refer HLD to know the request and responses of the GRPC mock service

• Based on proto files, you will be able to see the sample request of the service need to be mocked. example below. this is the sample request of DEBUG resolver which we are mocking pilot project

  {
   "ophid": "",
   "command": {
   "cmd": "",
   "args": [""],
   "env": [""],
   "workdir": ""
   },
  "stdin": "",
  "in": {
  	"namespace": "",
   "service": "",
   "index": 0,
   "container": ""
  }
  }
  

• Create the response based on the request. the response can be unary or stream response. based on the design of the service

• Create error use cases for all the possible GRPC error codes. While creating error use cases, keep one parameter mandatory in your request, based on the parameter create the error response. Example below

   ````	
   Request
   {
   "ophid": "N123450001"
   "command": {
   "cmd": "cat",
   "args": ["/var/named/named.conf"],
   "env": [],
    "workdir": ""
    },
    "stdin": ""
    }
   Response:
   Select different error code
   Ex: Server Unavailable 
   ```
  

3. Once the request response creation completed, integrate the Traffic parrot end point with actual service. In our case, Diagnostic service is the one depends on the debug resolver mock service.

4. Verify the sample request and responses

5. Creation Docker Image

   • With the proto and mapping files, Create the zip file of Traffic parrot
   • Create Docker Image and push to docker hub. (For pilot project, I have pushed this image to my private repo.

6. Follow the instructions to deploy **Traffic Parrot ** in k3d

   docker login <docker username>
   docker password
    export IMAGE=docker.io/prakashkb/trafficparrot:1.0.3       
     docker pull $IMAGE && k3d import-images $IMAGE
     kubectl apply -f deploy/trafficparrot.yaml
   

7. Install Diagnostic service - This actual code of Diagnostic service which is SUT (service under test)

   1. # import once only
      export IMAGE=infobloxcto/atlas.onprem.diagnostic-server:<Refer the latest version of image>
      docker pull $IMAGE && k3d import-images $IMAGE
      #Edit the deploy/diagnostic.yaml to confgiure debug resolver mock service
      debugresolver:
        enabled: true
            address: 10.120.250.248 <External IP of Traffic parrot instance> :5552 <Non TLS port> 
      kubectl apply -f deploy/diagnostic.yaml
      
      

   2. Test

      curl --request GET \
        --url http://localhost:8080/atlas-onprem-diagnostic-service/v1/remotecommands \