Performance Testing using Google cloud Build - K6
Performance Testing using Google cloud Build - K6
This tutorial shows you how to create a continuous delivery pipeline using Google Kubernetes Engine (GKE), Cloud Source Repositories and Cloud Build. After you create a sample app, you configure these services to automatically build, test, and deploy it. When you modify the app code, the changes trigger the continuous delivery pipeline to automatically rebuild, retest, and redeploy the new version.
This involves 3 stages :
Create a docker image and push it to the cloud repository
Let's start with - Create a docker image and push it to the cloud repository
You must have the google account to run the K6 tests using cloud build. Refer here
To interact with google cloud from the terminal, you need to interact with kubectl API servers. To make this work, need to install gcloud SDK. Refer here
Note : make sure you have completed both the above steps, to proceed further.
Once you have all the required SDK and google account, You need below files to get started
Dockerfile
cloudbuild.yaml
K6 Script
Get the above required files to be used for this exercise from here.
Clone the repo and open the folder in any editor.
Go through the Dockerfile and modify based on your requirement.
Create an image out of the dockerfile - Run below command in the root of repo.
docker build -t test:k6
The above command will create a base image against your Dockerfile.
This base image needs to push to the google cloud repository. Use below command to push it to the gcr.
docker ps -a - This will give you the image created.
docker tag <imagename> k6:latest - This will tag your base image with the name k6.
docker push gcr.io/<Your cloud ProjectName>/k6:latest
If you are unsure cloud project Name use below command. This will return your project name which you can enter in command.
gcloud projects list
Sample Output, Once you do docker push.
The push refers to repository [gcr.io/<Your cloud ProjectName>/k6]
a4974f6c41a8: Pushed
028a6b5f6661: Pushed
e57b0c314041: Layer already exists
dccc633e2921: Layer already exists
531743b7098c: Layer already exists
latest: digest: sha256:32e018d7b6915e6d948f638e77c073941722f1078f608c16af026ef3b6ccca01 size: 1365
If you want to run the test locally using the image. Run below command.
docker run --rm -p 1080:1080 gcr.io/<Your cloud ProjectName>/k6:latest
Your next action is to run this build using google cloud image. Let's proceed. If you do not want your test to be run with the deployment, skip all the below steps and Refer here.
You need to create a deployment with the image [referring to the image which is already pushed in the gcr]. Rub below command.
kubectl create deployment baselinetest --image=gcr.io/<Your cloud ProjectName>/k6:latest
Sample Output :
deployment.apps/baselinetest created
Create replica's of your pods.
kubectl scale deployment baselinetest --replicas=3
Sample Output :
deployment.apps/baselinetest scaled
Make autoscaling.
kubectl autoscale deployment baselinetest --cpu-percent=80 --min=1 --max=5
Sample Output :
horizontalpodautoscaler.autoscaling/baselinetest autoscaled
Check all the pods
kubectl get pods
Sample Output :
NAME READY STATUS RESTARTS AGE
baselinetest-75c4c4c8cb-p99qh 0/1 ImagePullBackOff 0 105s
baselinetest-75c4c4c8cb-sllbj 0/1 ErrImagePull 0 39s
baselinetest-75c4c4c8cb-wjxjr 0/1 ErrImagePull 0 39s
Now its time to integrate the cloud build with github. Refer here.