jmeter-load-testing

This repo used for auto scaling JMeter Agents with Kubernetes

1. Framework Requirements

• Core Features:
  • Automate the setup and teardown of JMeter agents.
  • Distribute load across multiple JMeter instances.
  • Monitor test progress and aggregate results.
  • Scalability using Kubernetes (K8s) for orchestration.
• Key Components:
  • JMeter Master: Coordinates test execution.
  • JMeter Agents: Run test scripts.
  • Result Aggregator: Collects and analyzes results.

2. Prerequisites

• Install Docker.
• Set up a Kubernetes cluster (Minikube) for local testing or a cloud provider like GKE, AKS, or EKS).
• Install kubectl and Helm.
• Install JMeter locally for testing scripts.

3. Create JMeter Docker Images

docker build -t your-repo/jmeter:latest .
docker push your-repo/jmeter:latest

4. Run Kubectl Context

copy your ./kubernetes/.kube/config to ~/.kube/config file

Run the context

 kubectl config use-context prod-cluster

Verify the ccurrent context

kubectl config current-context

List All contexts

kubectl config get-contexts

Restart your kubectl proxy (if applicable):

kubectl proxy --port=8080

Run before any other step

in powershell run as a administrator
 minikube start --driver=hyperv

5. Deploy Resources to Kubernetes

Validate yaml manually

kubectl apply --dry-run=client --validate=true -f jmeter-agent-deployment.yaml

Apply Configurations:

kubectl create namespace jmeter-ns

kubectl apply -f jmeter-agent-deployment.yaml
kubectl apply -f jmeter-master-deployment.yaml

kubectl apply -f jmeter-agent-service.yaml

Verify Deployments:

kubectl get pods

NAME                             READY   STATUS              RESTARTS   AGE
jmeter-agent-84bcbcb4f-7zzqf     0/1     ContainerCreating   0          4s
jmeter-agent-84bcbcb4f-cj6tp     0/1     ContainerCreating   0          4s
jmeter-agent-84bcbcb4f-ptlbd     0/1     ContainerCreating   0          4s
jmeter-master-57b5c764d9-hn4wc   0/1     ContainerCreating   0          52s

kubectl get services

NAME         TYPE        CLUSTER-IP   EXTERNAL-IP   PORT(S)   AGE
kubernetes   ClusterIP   10.96.0.1    <none>        443/TCP   15m

6. Run the test

Configure test parameters

kubectl exec -it jmeter-master -n jmeter-ns -- jmeter -n -t /tests/HTTPSignInSignOffPlan.jmx -Jusers=50 -Jduration=300 -R jmeter-agent-0, jmeter-agent-1, jmeter-agent-2

Run the test

kubectl logs -f deployment/jmeter-master -n jmeter

7. Collect and Aggregate Results

• Use Kubernetes logs to fetch results:

  kubectl cp jmeter/jmeter-master:/test/test-results.jtl ./test-results.jtl
  
  

Cleanup

kubectl delete namespace jmeter-ns

8. Monitor and Scale

• Use Kubernetes Horizontal Pod Autoscaler (HPA) to scale agents based on CPU usage:

kubectl autoscale deployment jmeter-agent --cpu-percent=50 --min=1 --max=10

9. Framework Structure

Organize the project:

jmeter-load-testing/
├── Dockerfile
├── kubernetes/
│   ├── jmeter-master-deployment.yaml
│   ├── jmeter-agent-deployment.yaml
│   └── configmap.yaml
├── helm-chart/
└── tests/
    ├── test.jmx
    ├── test-data.csv

**More trouble shooting techniques **

**couldn’t get current server API group list: Get “http://localhost:8080/api?timeout=32s”: dial tcp 127.0.0.1:8080: connect: connection refused **

ChatGPT said:

The error message indicates that a client application, likely kubectl, is trying to access the Kubernetes API server at http://localhost:8080, but the connection is being refused. This is typically caused by the API server not running, being configured incorrectly, or the client not being able to reach it. Here’s how to troubleshoot and resolve the issue:

---

1. Check Kubernetes Context and Configuration

• Run the following to inspect the current configuration:
• Ensure the current-context points to the correct cluster.
• To change the context, use:

---

2. Verify Kubernetes API Server is Running

• If you are running Kubernetes locally (e.g., via minikube or kind), ensure the cluster is started:

  • Minikube:

    If it’s not running, start it:

  • Kind:

    If no clusters are listed, create one:

---

3. Inspect API Server Endpoint

• The default API server address is unlikely to be http://localhost:8080 unless you’re running a local proxy (e.g., kubectl proxy).

• Check the correct endpoint in the kubeconfig:

  Look for the server entry under the current context, which should point to the correct API server URL.

---

4. Start kubectl proxy if Needed

• If you need to connect through http://localhost:8080, you might need to run kubectl proxy:
• This will forward requests from localhost:8080 to the API server.

---

5. Check Network Connectivity

• Ensure the client can reach the API server’s actual endpoint.
• Test the connection using:
• Replace <API_SERVER_ENDPOINT> with the actual API server URL (e.g., https://<your-cluster-endpoint>).

---

6. Restart Kubernetes Components

• If running Kubernetes locally, restarting components may help:
• If running on a managed cluster, verify the API server status with your cloud provider.

---

7. Check for Firewall/Proxy Issues

• If you’re on a corporate network, ensure there are no firewalls or proxies blocking access to the API server.

8. Troubleshoot for Minikube

$minikube start
* minikube v1.34.0 on Microsoft Windows 11 Home 10.0.22631.4602 Build 22631.4602
* Unable to pick a default driver. Here is what was considered, in preference order:
  - docker: Not healthy: "docker version --format -:" exit status 1: error during connect: Get "http://%2F%2F.%2Fpipe%2FdockerDesktopLinuxEngine/v1.47/version": open //./pipe/dockerDesktopLinuxEngine: The system cannot find the file specified.
  - docker: Suggestion:  <https://minikube.sigs.k8s.io/docs/drivers/docker/>
  - hyperv: Not healthy: Hyper-V requires Administrator privileges
  - hyperv: Suggestion: Right-click the PowerShell icon and select Run as Administrator to open PowerShell in elevated mode. <>
* Alternatively you could install one of these drivers:
  - virtualbox: Not installed: unable to find VBoxManage in $PATH
  - qemu2: Not installed: exec: "qemu-system-x86_64": executable file not found in %PATH%
  - podman: Not installed: exec: "podman": executable file not found in %PATH%

https://www.virtualbox.org/wiki/Downloads

https://www.qemu.org/download/

https://www.msys2.org/

https://podman.io/docs/installation

$kubectl get nodes
E1215 09:39:19.938225   20516 memcache.go:265] couldn't get current server API group list: Get "http://localhost:8080/api?timeout=32s": dial tcp [::1]:8080: connectex: No connection could be made because the target machine actively refused it.
https://discuss.kubernetes.io/t/couldnt-get-current-server-api-group-list-get-http-localhost-8080-api-timeout-32s-dial-tcp-127-0-0-1-connect-connection-refused/25471

Resolve by
KUBECONFIG=$HOME/.kube/config
sudo chown $(id -u):$(id -g) $HOME/.kube/config

if this still doesnt resolve. then,
All you need is to remove containerd service and reinstall it again


Install the latest PowerShell for new features and improvements! https://aka.ms/PSWindows

. : File C:\Users\srinivas\OneDrive\Documents\WindowsPowerShell\Microsoft.PowerShell_profile.ps1 cannot be loaded
because running scripts is disabled on this system. For more information, see about_Execution_Policies at
https:/go.microsoft.com/fwlink/?LinkID=135170.
At line:1 char:3
+ . 'C:\Users\srinivas\OneDrive\Documents\WindowsPowerShell\Microsoft.P ...
+   ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    + CategoryInfo          : SecurityError: (:) [], PSSecurityException
    + FullyQualifiedErrorId : UnauthorizedAccess
https://learn.microsoft.com/en-us/powershell/scripting/install/installing-powershell-on-windows?view=powershell-7.4#msi


powershell run as a administrator
 minikube start --driver=hyperv

E1215 10:18:45.156194 14160 status.go:263] The “minikube” host does not exist!

ChatGPT said:

The error indicates that Minikube is not able to find a cluster named minikube. This usually happens when no Minikube cluster has been created or the configuration is corrupted.

Here’s how to troubleshoot and resolve the issue:

---

1. Verify Existing Minikube Clusters

• Check if there are any existing Minikube clusters:
• If the list is empty or the minikube profile is missing, you need to create a new cluster.

---

2. Create a New Minikube Cluster

• Start a new Minikube cluster:
• Specify a profile name if needed:

---

3. Delete and Recreate the Cluster

• If there was an incomplete or corrupted setup, delete the existing configuration:
• Then, create a new cluster:

---

4. Check Virtualization

• Ensure your system supports virtualization, as Minikube often uses virtual machines.
• Verify virtualization is enabled:
  • Linux:
  • Windows (PowerShell):
  • macOS: Virtualization is enabled by default on modern Macs.
• If virtualization is not supported, you can use the --driver flag with alternatives like Docker:

---

5. Check the Minikube Driver

• Confirm the driver being used is properly installed and configured:
• Supported drivers include:
  • docker
  • virtualbox
  • hyperkit
  • none (bare metal)
• If needed, explicitly specify a driver:

---

6. Inspect Logs

• If issues persist, check Minikube logs for details:

---

7. Reset Minikube

• If your Minikube setup is corrupted, reset Minikube entirely:
• After resetting, recreate the cluster:

---

Let me know if you encounter any issues during these steps!