CECG Reference Application - GoLang

P2P Interface

The P2P interface is how the generated pipelines interact with the repo. For the CECG reference this follows the 3 musketeers pattern of using:

Make
Docker
Compose

These all need to be installed.

Structure

Service

Service source code, using Go.

Functional

Functional Tests using Cucumber Godog

Please refer to the docker compose file for instructions on how to do that.

NFT

Load tests using K6.

Running the application locally

Application

make run-local

This application is exposed locally on port 8080 as well as being available to the tests when run with make. This is as they are in the same docker network.

Functional Tests

make stubbed-functional

You should see:

Feature: Greeting
  It's polite to greet someone

  Scenario: hello world returns ok       # functional.feature:4
    Given a rest service                 # functional_test.go:15 -> aRestService
INFO[0000] Hitting endpoint http://localhost:8080       
    When I call the hello world endpoint # functional_test.go:20 -> iCallTheHelloWorldEndpoint
    Then an ok response is returned      # functional_test.go:32 -> anOkResponseIsReturned

1 scenarios (1 passed)
3 steps (3 passed)

Non-Functional Tests

make stubbed-nft

You should see:

     ✓ status was 200
     
     checks.........................: 100.00% ✓ 6590       ✗ 0    
     data_received..................: 850 kB  14 kB/s
     data_sent......................: 560 kB  9.2 kB/s
     http_req_blocked...............: avg=22.7µs  min=0s    med=4µs    max=2.9ms   p(90)=10µs   p(95)=29µs   
     http_req_connecting............: avg=11.99µs min=0s    med=0s     max=1.87ms  p(90)=0s     p(95)=0s     
     http_req_duration..............: avg=2.86ms  min=649µs med=2.02ms max=25.18ms p(90)=4.91ms p(95)=7.27ms 
       { expected_response:true }...: avg=2.86ms  min=649µs med=2.02ms max=25.18ms p(90)=4.91ms p(95)=7.27ms 
     http_req_failed................: 0.00%   ✓ 0          ✗ 6590 
     http_req_receiving.............: avg=40.32µs min=5µs   med=23µs   max=8.7ms   p(90)=63µs   p(95)=93.54µs
     http_req_sending...............: avg=32.3µs  min=2µs   med=13µs   max=5.17ms  p(90)=41µs   p(95)=74.54µs
     http_req_tls_handshaking.......: avg=0s      min=0s    med=0s     max=0s      p(90)=0s     p(95)=0s     
     http_req_waiting...............: avg=2.79ms  min=631µs med=1.96ms max=25ms    p(90)=4.81ms p(95)=7.12ms 
     http_reqs......................: 6590    108.135663/s
     iteration_duration.............: avg=1s      min=1s    med=1s     max=1.02s   p(90)=1s     p(95)=1.01s  
     iterations.....................: 6590    108.135663/s
     vus............................: 52      min=10       max=199
     vus_max........................: 200     min=200      max=200

Running in minikube

Prereqs

A minikube cluster i.e. you've run minikube start
- You need to enable the ingress addon by executing: minikube addons enable ingress and then follow the instructions from the output e.g. if on mac run minikube tunnel
Kubectl or use minikube kubectl
Godog binary installed and in your $PATH

Registries

You'll need a registry. Register for a Docker hub account and create a private registry e.g. savvasm1/reference-service-go-pub.

If using Docker Desktop then login

Update the Makefile registry variable with your newly created registry.

Pushing the image

make docker-build
make docker-push

Deploying the service

kubectl apply -f k8s-manifests/namespace.yml
kubectl apply -f k8s-manifests/deployment.yml

The service should be running:

kubectl get pods -n reference-service-showcase
NAME                                 READY   STATUS    RESTARTS   AGE
reference-service-7cff68d485-q8mw5   1/1     Running   0          142m

Deploy the ingress and service:

kubectl apply -f k8s-manifests/expose.yml

kubectl get ingress -n reference-service-showcase
NAME                CLASS   HOSTS   ADDRESS        PORTS   AGE
reference-service   nginx   *       192.168.49.2   80      144m

If on Linux or MacOS you can now access the service on the IP address (which is the minikube IP).

curl localhost/hello
Hello World!%

If this doesn't work ensure you followed the instructions when enabling the minikube ingress addon.

Run the functional tests against deployed application

This shows how you can run the same tests locally and on a deployed version.

cd into functional/godogs

SERVICE_ENDPOINT="http://localhost:8080/service" godog run

Run the non-functional tests against deployed application

cd into the root dir (core-reference-application-go)

SERVICE_ENDPOINT="http://localhost:8080/service" k6 run ./nft/ramp-up/test.js

nzacharia/ref-go