Opensearch

kubesphere/opensearch

Use when installing or configuring the OpenSearch extension for KubeSphere, which provides distributed search and analytics engine for storing logs, events, auditing, and notification history

3k tokens

context cost

the whole folder, loaded on every use

files

instructions only

copies elsewhere

how many repositories repackaged it

17017

stars on the repo

on the repository, not the skill itself

Install

one command, takes just this skill from the repository

npx skills add https://github.com/kubesphere/kubesphere --skill opensearch

The instruction itself

29 sections, as written by the author

OpenSearch Extension

Overview

OpenSearch is a distributed search and analytics engine built into the KubeSphere WizTelemetry Observability Platform. It is used to store, search, and analyze observability data including logs, auditing events, K8s events, and notification history.

When to Use

Installing OpenSearch cluster for KubeSphere observability extensions
Configuring OpenSearch storage for logging, events, auditing, and notifications
Managing OpenSearch Dashboard and Curator components

Components

| Component | Description | Default |

|-----------|-------------|---------|

| opensearch-master | Master node for cluster coordination | 1 replica |

| opensearch-data | Data nodes for storing indices | 3 replicas |

| opensearch-dashboards | Web UI for visualizing data | disabled |

| opensearch-curator | Scheduled task to clean old indices | enabled |

Prerequisites

Check Installation Status

# Check if OpenSearch is installed
kubectl get installplan opensearch -o jsonpath='{.spec.enabled}'

# Get installed version
kubectl get extension opensearch -o jsonpath='{.status.installedVersion}'

# Get target clusters
kubectl get installplan opensearch -o jsonpath='{.spec.clusterScheduling.placement.clusters}'

Returns:

"true" - installed and enabled
"false" - installed but disabled
Empty/Error - not installed

Get Available Clusters

kubectl get clusters -o jsonpath='{.items[*].metadata.name}'

Confirm Target Clusters (MUST DO)

⚠️ CRITICAL: Do Not guess.

⚠️ CRITICAL: DO NOT proceed until target clusters are determined.

If user explicitly specified target clusters in the request → Use those clusters directly
If user did NOT specify target clusters → You MUST ask user to confirm which clusters to deploy to

Ask user (if not specified):

Available clusters: host, dev
Which clusters do you want to deploy OpenSearch to?

Get Latest Version

MUST do this to get the latest version:

kubectl get extensionversions -l kubesphere.io/extension-ref=opensearch -o jsonpath='{range .items[*]}{.spec.version}{"\n"}{end}' | sort -V | tail -1

Installation

⚠️ CRITICAL: InstallPlan metadata.name MUST be opensearch. DO NOT use any other name.

Use the latest version obtained from "Get Latest Version" step.

Minimal Installation

apiVersion: kubesphere.io/v1alpha1
kind: InstallPlan
metadata:
  name: opensearch
spec:
  extension:
    name: opensearch
    version: <VERSION>  # From Get Latest Version step
  enabled: true
  upgradeStrategy: Manual
  clusterScheduling:
    placement:
      clusters:
        - host

With OpenSearch Dashboard

apiVersion: kubesphere.io/v1alpha1
kind: InstallPlan
metadata:
  name: opensearch
spec:
  extension:
    name: opensearch
    version: <VERSION>  # From Get Latest Version step
  enabled: true
  upgradeStrategy: Manual
  config: |
    opensearch-dashboards:
      enabled: true
  clusterScheduling:
    placement:
      clusters:
        - host

Disable Curator

apiVersion: kubesphere.io/v1alpha1
kind: InstallPlan
metadata:
  name: opensearch
spec:
  extension:
    name: opensearch
    version: <VERSION>  # From Get Latest Version step
  enabled: true
  upgradeStrategy: Manual
  config: |
    opensearch-curator:
      enabled: false
  clusterScheduling:
    placement:
      clusters:
        - host

Configuration

Parameters

|-----------|------|---------|-------------|

| opensearch-master.replicas | int | 1 | Number of master nodes |

| opensearch-data.replicas | int | 3 | Number of data nodes |

| opensearch-data.service.nodePort | int | 30920 | NodePort for external access |

Resource Configuration

opensearch-master:
  replicas: 1
  opensearchJavaOpts: "-Xmx512M -Xms512M"
  resources:
    requests:
      cpu: "100m"
      memory: "512Mi"
    limits:
      cpu: "500m"
      memory: "1Gi"

opensearch-data:
  replicas: 3
  opensearchJavaOpts: "-Xmx2048M -Xms2048M"
  resources:
    requests:
      cpu: "200m"
      memory: "2Gi"
    limits:
      cpu: "2000m"
      memory: "4Gi"
  persistence:
    size: 50Gi
    storageClass: "local-volume"

Operations

Access Target Cluster

> ⚠️ CRITICAL: You MUST access the target cluster BEFORE checking status or running any kubectl commands below.

> ❌ DO NOT run kubectl commands without using the target cluster kubeconfig

> ✅ MUST run: kubectl --kubeconfig=/tmp/<cluster>-kubeconfig ...

Checklist (must complete before any operation):

[ ] 1. Find target clusters:

  kubectl get installplan opensearch -o jsonpath='{.spec.clusterScheduling.placement.clusters}'

[ ] 2. Get cluster kubeconfig:

  kubectl get cluster <cluster-name> -o jsonpath='{.spec.connection.kubeconfig}' | base64 -d > /tmp/<cluster-name>-kubeconfig

[ ] 3. Use this kubeconfig for ALL subsequent commands

> Note: Replace <cluster-name> with the actual cluster name. Use spec.connection.kubeconfig for imported clusters.

Check Status

> ⚠️ REQUIRED: Complete Access Target Cluster checklist FIRST

# Extension status
kubectl --kubeconfig=/tmp/<cluster>-kubeconfig get extension opensearch

# InstallPlan status
kubectl --kubeconfig=/tmp/<cluster>-kubeconfig get installplan opensearch

Get Pods

> ⚠️ REQUIRED: Complete Access Target Cluster checklist FIRST

# All pods
kubectl --kubeconfig=/tmp/<cluster>-kubeconfig get pods -n kubesphere-logging-system -l app.kubernetes.io/instance=opensearch-agent

# By component
kubectl --kubeconfig=/tmp/<cluster>-kubeconfig get pods -n kubesphere-logging-system -l app.kubernetes.io/name=opensearch-master
kubectl --kubeconfig=/tmp/<cluster>-kubeconfig get pods -n kubesphere-logging-system -l app.kubernetes.io/name=opensearch-data

# Status
kubectl --kubeconfig=/tmp/<cluster>-kubeconfig get pods -n kubesphere-logging-system -l app.kubernetes.io/instance=opensearch-agent -o wide

External Endpoint

> ⚠️ REQUIRED: Complete Access Target Cluster checklist FIRST

OpenSearch uses NodePort (default: 30920) to expose service externally.

# Get NodePort service
kubectl --kubeconfig=/tmp/<cluster>-kubeconfig get svc -n kubesphere-logging-system opensearch-cluster-data

# Get node IP
kubectl --kubeconfig=/tmp/<cluster>-kubeconfig get nodes -o jsonpath='{.items[0].status.addresses[?(@.type=="InternalIP")].address}'

# Access OpenSearch externally
# URL: https://<node-ip>:30920
curl -k -u admin:admin "https://<node-ip>:30920/_cluster/health"

Output Contract (For Other Skills)

> ⚠️ This section defines what information this skill provides to other skills

When OpenSearch is successfully deployed, other skills can retrieve:

| Field | Example Value | Description |

|-------|---------------|-------------|

| endpoint | https://127.0.0.1:30920 | Full URL for Vector sink |

| nodePort | 30920 | NodePort number |

| nodeIP | 127.0.0.1 | Node internal IP |

| auth.user | admin | Default username |

| auth.password | admin | Default password |

How other skills should get this info:

Find which cluster OpenSearch was deployed to:

   kubectl get installplan opensearch -o jsonpath='{.spec.clusterScheduling.placement.clusters}'

Get cluster kubeconfig:

   kubectl get cluster <cluster-name> -o jsonpath='{.spec.connection.kubeconfig}' | base64 -d > /tmp/<cluster-name>-kubeconfig

Get OpenSearch endpoint:

   # NodePort
   NODE_PORT=$(kubectl --kubeconfig=/tmp/<cluster>-kubeconfig get svc -n kubesphere-logging-system opensearch-cluster-data -o jsonpath='{.spec.ports[?(@.port==9200)].nodePort}')
   
   # Node IP
   NODE_IP=$(kubectl --kubeconfig=/tmp/<cluster>-kubeconfig get nodes -o jsonpath='{.items[0].status.addresses[?(@.type=="InternalIP")].address}')
   
   # Full endpoint
   echo "https://${NODE_IP}:${NODE_PORT}"

Update Configuration

apiVersion: kubesphere.io/v1alpha1
kind: InstallPlan
metadata:
  name: opensearch
spec:
  extension:
    name: opensearch
    version: <VERSION>  # From Get Latest Version step
  enabled: true
  upgradeStrategy: Manual
  config: |
    opensearch-data:
      replicas: 5
      opensearchJavaOpts: "-Xmx4096M -Xms4096M"
    opensearch-dashboards:
      enabled: true
  clusterScheduling:
    placement:
      clusters:
        - host

Uninstall

Uninstall from all clusters:

kubectl delete installplan opensearch

Uninstall from specific cluster:

To remove OpenSearch from a specific cluster, update the InstallPlan by removing that cluster from clusterScheduling.placement.clusters:

apiVersion: kubesphere.io/v1alpha1
kind: InstallPlan
metadata:
  name: opensearch
spec:
  extension:
    name: opensearch
    version: <VERSION>
  enabled: true
  upgradeStrategy: Manual
  config: |
    opensearch-data:
      replicas: 3
  clusterScheduling:
    placement:
      clusters:
        - <REMAINING_CLUSTERS>  # Remove the cluster you want to uninstall from

Troubleshooting

> ⚠️ REQUIRED: Complete Access Target Cluster checklist FIRST

Pod Issues

kubectl --kubeconfig=/tmp/<cluster>-kubeconfig get pods -n kubesphere-logging-system -l app.kubernetes.io/instance=opensearch-agent

# Pod events
kubectl --kubeconfig=/tmp/<cluster>-kubeconfig describe pods -n kubesphere-logging-system -l app.kubernetes.io/instance=opensearch-agent

OpenSearch Health

> ⚠️ REQUIRED: Complete Access Target Cluster checklist FIRST

# Cluster health
curl -k -u admin:admin "https://<node-ip>:30920/_cluster/health"

# List indices
curl -k -u admin:admin "https://<node-ip>:30920/_cat/indices"

Common Issues

| Issue | Solution |

|-------|----------|

| Pods not starting | Check node resources and storage availability |

| Out of memory | Increase JVM heap size |

| Cannot connect | Check NodePort firewall rules |

| Index storage full | Increase PVC size or enable Curator |

Notes

Resource Requirements: OpenSearch requires significant CPU and memory
Storage: Data nodes use persistent storage
NodePort: Default port is 30920
Multi-cluster: Can be deployed to specific clusters

How to use it

Copy the folder

Take kubesphere/opensearch from the repository into ~/.claude/skills for personal use, or into .claude/skills inside a project.

Check the name does not clash

The agent identifies a skill by the name field in its header. Two skills with the same name cannot sit side by side — one of them will be ignored.