Run a load test with k6

Home
Testing

Last updated: 20 Nov 2024

These instructions are based on this blog post.

Determine the name of the Kubernetes context to use:
```
kubectl config get-contexts
```
The output will list the environments and their names (e.g. integration, staging or production, or however they are named in your ~/.kube/config file).
Log into AWS and set the context to the environment you wish to test (this is the context name from the previous step):
```
ENVIRONMENT=staging
ROLE=admin

eval $(gds aws govuk-${ENVIRONMENT}-${ROLE} -e --art 8h)
kubectl config use-context $ENVIRONMENT
```
You will need to assume the admin role in order to deploy k6 to the cluster.
Clone the k6-operator repo and make this:
```
git clone git@github.com:grafana/k6-operator.git && cd k6-operator
make deploy
```
If you get an error similar to invalid go version '1.22.0': must match format 1.23 then you need to upgrade go to the correct version. This can be done by using Homebrew, e.g. brew install go@1.23

Write your test script, for example to test 3000 virtual users against the Collections app’s /world page over 30 seconds (bypassing router):

import http from 'k6/http';
import { check } from 'k6';

export const options = {
  vus: 3000,
  duration: '30s',
}

export default function() {
  const res = http.get('http://collections.apps.svc.cluster.local/world');

  check(res, {
    'is status 200': (r) => r.status === 200,
    'is status 2xx, not 200': (r) => r.status >= 201 && r.status <= 299,
    'is status 3xx': (r) => r.status >= 300 && r.status <= 399,
    'is status 4xx': (r) => r.status >= 400 && r.status <= 499,
    'is status 5xx': (r) => r.status >= 500 && r.status <= 599,
  });
}

Run your code locally to make sure it works:
```
k6 run test.js
```
You may see errors related to connections as you may not be able to establish the correct number of connections from your machine.
Deploy the test script:
```
kubectl create configmap my-test-name --from-file /path/to/our/test.js
```
Replace my-test-name with a name for this load test. The name will be used later on, e.g. in the name of the pods that run the test.

Create a YAML file to contain your custom resource:

apiVersion: k6.io/v1alpha1
kind: K6
metadata:
  name: k6-load-test
spec:
  parallelism: 4
  script:
    configMap:
      name: my-test-name
      file: test.js

This resource will spawn 4 test runners. You may need to increase this value to run a large scale test.

Deploy your custom resource:
```
kubectl apply -f /path/to/our/k6/custom-resource.yml
```
At this point, your test will have started. You can track progress using kubectl get pods.
Once all the pods are marked as completed, you can retrieve the output of each runner by requesting the logs.

First get the names of each pod:
```
kubectl get pods
```
For each runner pod, retrieve the logs:
```
kubectl logs my-test-name-1-abcde
```
After running the tests, tidy up by deleting your custom resource:
```
kubectl delete -f /path/to/our/k6/custom-resource.yml
```

Run a load test with k6

More in the Testing section

Learn

How to...