Integration test

Create file integration-test.sh

content_copysleep 5s
PORT=$(kubectl -n default get svc ${serviceName} -o json | jq .spec.ports[].nodePort)
echo $PORT
echo $applicationURL:$PORT/$applicationURI
if [[ ! -z "$PORT" ]];
then
    response=$(curl -s $applicationURL:$PORT$applicationURI)
    http_code=$(curl -s -o /dev/null -w "%{http_code}" $applicationURL:$PORT$applicationURI)
    if [[ "$response" == 100 ]];
        then
            echo "Increment Test Passed"
        else
            echo "Increment Test Failed"
            exit 1;
    fi;
    if [[ "$http_code" == 200 ]];
        then
            echo "HTTP Status Code Test Passed"
        else
            echo "HTTP Status code is not 200"
            exit 1;
    fi;
else
        echo "The Service does not have a NodePort"
        exit 1;
fi;

The code performs automated testing on a service running on a Kubernetes cluster. Here's what each section of the script does:

Retrieve the NodePort for a Kubernetes service specified by $serviceName:

PORT=$(kubectl -n default get svc ${serviceName} -o json | jq .spec.ports[].nodePort)

This command uses kubectl to retrieve the service's details in JSON format, and jq to extract the NodePort value from the JSON output. The NodePort is stored in the $PORT variable.

Output the service's URL: echo $applicationURL:$PORT/$applicationURI

This command outputs the URL that the service can be accessed at, constructed by combining the $applicationURL, $PORT, and $applicationURI variables.

Send HTTP requests to the service and perform tests:

content_copyif [[ ! -z "$PORT" ]];
then
    response=$(curl -s $applicationURL:$PORT$applicationURI)
    http_code=$(curl -s -o /dev/null -w "%{http_code}" $applicationURL:$PORT$applicationURI) 
    if [[ "$response" == 100 ]];
        then
            echo "Increment Test Passed"
        else
            echo "Increment Test Failed"
            exit 1;
    fi;
    if [[ "$http_code" == 200 ]];
        then
            echo "HTTP Status Code Test Passed"
        else
            echo "HTTP Status code is not 200"
            exit 1;
    fi;
else
    echo "The Service does not have a NodePort"
    exit 1;
fi;

In this section, two HTTP requests are sentto the service. The first request retrieves the response body and stores it in the $response variable.

Since this application has function +1 value per request, and now we have applicationURL="/increment/99". So we need "$response" == 100 in code. Ifthe response body is "100", the "Increment Test Passed" message is printed to the console.

The second request retrieves the HTTP response code and stores it in the $http_code variable. The two if statements then perform tests on these values. If the HTTP response code is "200", the "HTTP Status Code Test Passed" message is printed. If either of these tests fail, an error message is printed and the script exits with a non-zero status code. If the $PORT variable is empty, an error message is printed and the script exits with a non-zero status code.

Test application:

content_copycurl -s -o /dev/null -w "%{http_code}" $applicationURL:$PORT$applicationURI

This is a curl command that retrieves the HTTP status code of a URL's response without printing the response body to the console. Here's what each piece of the command does:

curl: the command that makes HTTP requests to web servers.

-s: a parameter that tells curl to operate in silent mode, meaning that it won't display any progress information or error messages in the console. It's useful for automated testing since it avoids cluttering the output.

-o /dev/null: a parameter that redirects the output of the command to /dev/null, a special file that discards all output. This means that the body of the response is not returned to the script, which would interfere with the tests.

-w "%{http_code}": a parameter that outputs only the HTTP status code, using a predefined format. "%"{http_code} is a curl variable that represents the HTTP status code of the response.

$applicationURL:$PORT$applicationURI: a dynamic URL that combines the $applicationURL, $PORT, and $applicationURI variables that the script has defined earlier. $applicationURL represents the base URL of the service, $PORT the NodePort number of the service, and $applicationURI a specific endpoint or resource path of the service.

Together, these parameters tell curl to send anHTTP request to the constructed URL, retrieve only the HTTP status code from the response headers, and discard the response body. The response status code is then assigned to the $http_code variable so that it can be used later in the script for testingpurposes to verify that the service is functioning as expected.

Create pipeline Integration Tests and build pipeline.

content_copystage('Integration Tests - DEV') {
      steps {
        script {
          try {
            withKubeConfig([credentialsId: 'kubeconfig']) {
              sh "bash integration-test.sh"
            }
          } catch (e) {
            withKubeConfig([credentialsId: 'kubeconfig']) {
              sh "kubectl -n default rollout undo deploy ${deploymentName}"
            }
            throw e
          }
        }
      }
    }