Merge pull request #28 from dinudaniel/feature/chaos-mesh-on-oke

[FEATURE] Add Chaos Mesh on OKE

Author: adinan-tech

chaos-mesh-on-oke-sample/README.md

@@ -0,0 +1,202 @@
+# Deploy Chaos Mesh on OKE using Oracle Resource Manager
+
+## Introduction
+
+Oracle Kubernetes Engine (OKE) is the managed Kubernetes service on Oracle Cloud Infrastructure (OCI). It provides a managed Kubernetes control plane, worker node pools, and native integration with OCI networking and load balancing.
+
+Chaos Mesh is an open source chaos engineering platform for Kubernetes. It lets you run controlled experiments against Kubernetes workloads and observe how applications behave under failures or resource pressure.
+
+This Terraform stack deploys an OKE cluster and installs Chaos Mesh on it. The included basic usage example deploys `nginx`, configures an HPA, and runs a Chaos Mesh `StressChaos` memory experiment.
+
+## Objectives
+
+- Deploy an enhanced OKE cluster and the required OCI networking resources.
+- Install Chaos Mesh on OKE using Oracle Resource Manager (ORM).
+- Show a basic usage example with an `nginx` deployment, HPA, and `StressChaos` experiment.
+
+## Prerequisites
+
+- Access to an active OCI tenancy.
+- A compartment where the resources will be created.
+- Sufficient privileges to create and manage Resource Manager stacks/jobs, OKE, networking, compute, block volume, and load balancer resources. Example policies:
+
+  ```text
+  allow group <group-name> to manage orm-stacks in compartment <compartment-name>
+  allow group <group-name> to manage orm-jobs in compartment <compartment-name>
+  allow group <group-name> to manage cluster-family in compartment <compartment-name>
+  ```
+
+- A machine with OCI CLI configured and `kubectl` installed.
+- SSH public key. The ORM stack schema requires this input.
+
+## Task 1: Deploy Chaos Mesh on OKE using Oracle Resource Manager
+
+1. Download the GitHub release archive as a `.zip` file.
+
+2. Unzip the archive on your local machine.
+
+3. In the OCI Console, create a Resource Manager stack:
+
+   - Open the navigation menu and go to **Developer Services** > **Resource Manager** > **Stacks**.
+   - Click **Create stack**.
+   - Select **My configuration**.
+   - In the **My configuration** section, select **Folder**.
+   - Upload the directory from the unzipped archive.
+   - Give the stack a meaningful name, for example `chaos-mesh-on-oke`.
+   - Choose the compartment where the stack metadata should be stored.
+   - Click **Next**.
+
+4. Configure stack variables:
+
+   - Choose the compartment where the OKE and networking resources will be created.
+   - Select the Kubernetes version.
+   - Provide your SSH public key.
+   - Keep **Enable Metrics Server** and **Enable cert-manager** enabled. They are required for the HPA example.
+   - Choose the Chaos Mesh dashboard service type. `ClusterIP` is recommended.
+   - Click **Next**.
+
+5. Review the stack configuration.
+
+6. Select **Run apply** and click **Create**.
+
+7. Wait for the apply job to finish. The stack creates:
+
+   - OCI networking resources: VCN, subnets, gateways, route/security resources, and network security groups.
+   - An enhanced OKE cluster named `chaosmesh`.
+   - A 3-node OKE node pool named `chaosmesh-pool`.
+   - OKE add-ons: cert-manager and Metrics Server.
+   - A Kubernetes namespace named `chaos-mesh`.
+   - Chaos Mesh resources, via helm, installed in `chaos-mesh` namespace.
+
+## Task 2: Access the OKE cluster
+
+1. In the OCI Console, open **Developer Services** > **Kubernetes Clusters (OKE)**.
+
+2. Select the cluster named `chaosmesh`.
+
+3. Click **Actions** drop-down menu and then click **Access Cluster**.
+
+4. Copy the OCI CLI command shown in the console.
+
+5. Paste and run the command in a shell on a machine where OCI CLI and kubectl are configured.
+
+6. Confirm that `kubectl` can access the cluster:
+
+   ```bash
+   kubectl get nodes
+   ```
+
+## Task 3: Access the Chaos Mesh web interface
+
+1. In the same shell, start port forwarding and leave this command running:
+
+   ```bash
+   kubectl port-forward -n chaos-mesh svc/chaos-dashboard 2333:2333
+   ```
+
+2. Open the Chaos Mesh dashboard in a browser:
+
+   ```text
+   http://localhost:2333
+   ```
+
+3. In another shell, generate a login token:
+
+   ```bash
+   kubectl create token chaos-dashboard -n chaos-mesh
+   ```
+
+4. In the web interface, give the token a name, for example `chaos`, and paste the token value generated above. Then click **Submit**
+
+## Task 4: Run a basic usage example with nginx, HPA, and StressChaos
+
+The example uses the Kubernetes manifests included in **chaos-mesh-on-oke-sample** directory:
+
+- `deployment.yaml` deploys `nginx`.
+- `hpa.yaml` creates the HPA. This is the Horizontal Pod Autoscaler (HPA) manifest.
+- `stresschaos.yaml` creates the Chaos Mesh memory stress experiment.
+
+1. Open a new shell and change to the Terraform directory from the unzipped archive:
+
+   ```bash
+   cd <unzipped-archive>
+   ```
+
+2. Create the `nginx` deployment:
+
+   ```bash
+   kubectl apply -f deployment.yaml
+   kubectl rollout status deployment/nginx -n default
+   kubectl get pods -n default -l app=nginx
+   ```
+
+3. Configure the HPA:
+
+   ```bash
+   kubectl apply -f hpa.yaml
+   kubectl get hpa nginx-hpa -n default
+   ```
+
+4. Create the Chaos Mesh stress experiment:
+
+   ```bash
+   kubectl apply -f stresschaos.yaml
+   kubectl get stresschaos nginx-mem-1 -n default
+   ```
+
+5. Observe the deployment while the experiment runs:
+
+   ```bash
+   kubectl get deployment nginx -n default -w
+   ```
+
+6. In another shell, observe the HPA:
+
+   ```bash
+   kubectl get hpa nginx-hpa -n default -w
+   ```
+
+7. Optional: watch the nginx pods directly:
+
+   ```bash
+   kubectl get pods -n default -l app=nginx -w
+   ```
+
+8. Expected behavior:
+
+   - Chaos Mesh injects memory stress into the `nginx` pod for the duration defined in `stresschaos.yaml`.
+   - The HPA detects increased memory consumption.
+   - The number of `nginx` pods increases.
+   - After the experiment ends and the HPA stabilization period passes, the deployment scales down again.
+
+## Task 5: Clean up
+
+1. Delete the Chaos Mesh stress experiment:
+
+   ```bash
+   kubectl delete -f stresschaos.yaml --ignore-not-found=true
+   ```
+
+2. Delete the HPA:
+
+   ```bash
+   kubectl delete -f hpa.yaml --ignore-not-found=true
+   ```
+
+3. Delete the `nginx` deployment:
+
+   ```bash
+   kubectl delete -f deployment.yaml --ignore-not-found=true
+   ```
+
+4. Stop the Chaos Mesh dashboard port-forward command with `Ctrl+C`.
+
+5. Destroy the OCI resources created by the Terraform stack:
+
+   - Go to **Developer Services** > **Resource Manager** > **Stacks**.
+   - Select the stack you created.
+   - Click **Destroy**.
+   - Review the destroy job and confirm.
+
+6. Wait for the destroy job to complete successfully.
+

chaos-mesh-on-oke-sample/addons.tf

@@ -0,0 +1,20 @@
+# Copyright (c) 2024, 2026, Oracle and/or its affiliates. All rights reserved.
+# The Universal Permissive License (UPL), Version 1.0 as shown at https://oss.oracle.com/licenses/upl/
+
+resource "oci_containerengine_addon" "metrics_server" {
+  count                            = var.enable_metrics_server ? 1 : 0
+  cluster_id                       = oci_containerengine_cluster.chaosmesh.id
+  addon_name                       = "KubernetesMetricsServer"
+  remove_addon_resources_on_delete = true
+
+  depends_on = [oci_containerengine_node_pool.chaosmesh-pool, oci_containerengine_addon.cert_manager]
+}
+
+resource "oci_containerengine_addon" "cert_manager" {
+  count                            = var.enable_cert_manager ? 1 : 0
+  cluster_id                       = oci_containerengine_cluster.chaosmesh.id
+  addon_name                       = "CertManager"
+  remove_addon_resources_on_delete = true
+
+  depends_on = [oci_containerengine_node_pool.chaosmesh-pool]
+}

chaos-mesh-on-oke-sample/cloud-init.sh

@@ -0,0 +1,11 @@
+#!/bin/bash
+
+# Copyright (c) 2024, 2026, Oracle and/or its affiliates. All rights reserved.
+# The Universal Permissive License (UPL), Version 1.0 as shown at https://oss.oracle.com/licenses/upl/
+
+mkdir -p /etc/crio/crio.conf.d
+echo "[crio]" > /etc/crio/crio.conf.d/11-default.conf
+echo "  [crio.image]" >> /etc/crio/crio.conf.d/11-default.conf
+echo "    short_name_mode=\"disabled\"" >> /etc/crio/crio.conf.d/11-default.conf
+curl --fail -H "Authorization: Bearer Oracle" -L0 http://169.254.169.254/opc/v2/instance/metadata/oke_init_script | base64 --decode >/var/run/oke-init.sh
+sudo bash /var/run/oke-init.sh

chaos-mesh-on-oke-sample/deployment.yaml

@@ -0,0 +1,23 @@
+# Copyright (c) 2024, 2026, Oracle and/or its affiliates. All rights reserved.
+# The Universal Permissive License (UPL), Version 1.0 as shown at https://oss.oracle.com/licenses/upl/
+
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: nginx
+  namespace: default
+spec:
+  replicas: 1
+  selector:
+    matchLabels:
+      app: nginx
+  template:
+    metadata:
+      labels:
+        app: nginx
+    spec:
+      containers:
+        - name: nginx
+          image: docker.io/nginx
+          ports:
+            - containerPort: 80

chaos-mesh-on-oke-sample/hpa.yaml

@@ -0,0 +1,22 @@
+# Copyright (c) 2024, 2026, Oracle and/or its affiliates. All rights reserved.
+# The Universal Permissive License (UPL), Version 1.0 as shown at https://oss.oracle.com/licenses/upl/
+
+apiVersion: autoscaling/v2
+kind: HorizontalPodAutoscaler
+metadata:
+  name: nginx-hpa
+  namespace: default
+spec:
+  scaleTargetRef:
+    apiVersion: apps/v1
+    kind: Deployment
+    name: nginx
+  minReplicas: 1
+  maxReplicas: 10
+  metrics:
+  - type: Resource
+    resource:
+      name: memory
+      target:
+        type: AverageValue
+        averageValue: 100Mi