Skip to content

Latest commit

 

History

History
68 lines (54 loc) · 1.91 KB

README.MD

File metadata and controls

68 lines (54 loc) · 1.91 KB
Raw

Camunda-Benchmark-Operator 🏋️‍♀️

An operator that helps you perform benchmarks.

Your first benchmark

This requires that you know how to run the operator, you might be able to figure that out from the config folder.

In order to benchmark you need to create a secret with the credentials for your zeebe cluster.

An example secret is depicted below. Make sure to use the same keys in stringData. If you change the name you also have to adjust the "credentialsSecretName" in the benchmark definition.

apiVersion: v1
kind: Secret
metadata:
  name: cloud-credentials
type: Opaque
stringData:
  clientId: <your-client-id>
  clientSecret: <your-client-secret>
  zeebeAddress: <your-zeebe-address-with-port>
  authServer: <your-auth-server-url>

To now create a benchmark you apply the following manifest to your cluster. Make sure that the name of the secret matches with the secret you created earlier.

apiVersion: cloud.camunda.io/v1alpha1
kind: Benchmark
metadata:
  name: benchmark-sample
spec:
  credentialsSecretName: cloud-credentials
  workerCount: 2
  processStarterRate: 200
  starterReplicas: 3
  duration: 1m30s

Current Status

Run the following command if you want to get more insights on the current status of the benchmark: kubectl describe benchmark benchmark-sample

Spec:
  Credentials Secret Name:  cloud-credentials
  Duration:                 2m
  Process Starter Rate:     200
  Starter Replicas:         3
  Worker Count:             2
Status:
  Progress:    Running
  Start Time:  2022-02-11T21:34:21Z

Events:        <none>

You can also run kubectl get bench benchmark-sample and it will print this:

NAME                    # WORKERS   # STARTERS   STARTER RATE   DURATION   AGE    PROGRESS
benchmark-sample-gets   2           1            200            2m         106s   Running

In the future this should also log the results of the benchmark.