Integration

Integrate K8sGPT into your CI/CD pipelines to catch issues before they reach production:

GitHub Actions

name: K8sGPT Analysis
on:
  push:
    branches: [ main ]
  pull_request:
    branches: [ main ]

jobs:
  analyze:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v3
      - name: Install K8sGPT
        run: |
          curl -sSfL https://raw.githubusercontent.com/k8sgpt-ai/k8sgpt/main/install.sh | sh
      - name: Run Analysis
        run: |
          k8sgpt analyze --namespace=default
        env:
          K8SGPT_OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
          KUBECONFIG: ${{ secrets.KUBECONFIG }}

GitLab CI

analyze:
  image: ubuntu:latest
  before_script:
    - apt-get update && apt-get install -y curl
    - curl -sSfL https://raw.githubusercontent.com/k8sgpt-ai/k8sgpt/main/install.sh | sh
  script:
    - k8sgpt analyze --namespace=default
  environment:
    name: production
  variables:
    K8SGPT_OPENAI_API_KEY: "$OPENAI_API_KEY"
    KUBECONFIG: "$KUBECONFIG"

Connect K8sGPT with your monitoring stack:

Prometheus

apiVersion: monitoring.coreos.com/v1
kind: ServiceMonitor
metadata:
  name: k8sgpt
  labels:
    release: prometheus
spec:
  selector:
    matchLabels:
      app.kubernetes.io/name: k8sgpt
  endpoints:
  - port: metrics
    interval: 30s

Grafana Dashboard

Import the K8sGPT dashboard into Grafana:

# Dashboard ID: 12345
# Available at: https://grafana.com/grafana/dashboards/12345

Send K8sGPT analysis results to various notification systems:

Slack

apiVersion: v1
kind: Secret
metadata:
  name: k8sgpt-slack-secret
type: Opaque
stringData:
  SLACK_TOKEN: "xoxb-your-token"
  SLACK_CHANNEL: "#kubernetes-alerts"

Email

apiVersion: v1
kind: ConfigMap
metadata:
  name: k8sgpt-email-config
data:
  SMTP_SERVER: "smtp.example.com"
  SMTP_PORT: "587"
  FROM_EMAIL: "k8sgpt@example.com"
  TO_EMAIL: "team@example.com"

Webhooks

apiVersion: v1
kind: ConfigMap
metadata:
  name: k8sgpt-webhook-config
data:
  WEBHOOK_URL: "https://api.example.com/webhook"
  WEBHOOK_METHOD: "POST"
  WEBHOOK_HEADERS: '{"Content-Type": "application/json"}'

Integrate K8sGPT with service mesh solutions:

Istio

apiVersion: networking.istio.io/v1alpha3
kind: VirtualService
metadata:
  name: k8sgpt
spec:
  hosts:
  - k8sgpt
  http:
  - route:
    - destination:
        host: k8sgpt
        port:
          number: 8080

Linkerd

apiVersion: linkerd.io/v1alpha2
kind: ServiceProfile
metadata:
  name: k8sgpt
  namespace: default
spec:
  routes:
  - name: k8sgpt-api
    condition:
      pathRegex: "/api/.*"
    isRetryable: true
    timeout: 5000ms

Connect K8sGPT with your logging infrastructure:

Elasticsearch

apiVersion: v1
kind: ConfigMap
metadata:
  name: k8sgpt-elasticsearch-config
data:
  ELASTICSEARCH_URL: "http://elasticsearch:9200"
  ELASTICSEARCH_INDEX: "k8sgpt-analysis"

Fluentd

<match k8sgpt.**>
  @type elasticsearch
  host elasticsearch
  port 9200
  logstash_format true
  logstash_prefix k8sgpt
  <buffer>
    @type memory
    flush_interval 5s
  </buffer>
</match>

Create custom integrations using K8sGPT's API:

REST API

# Start the API server
k8sgpt serve --port=8080

# Example API call
curl -X POST http://localhost:8080/api/analyze \
  -H "Content-Type: application/json" \
  -d '{"namespace": "default", "analyzers": ["pod"]}'

Webhook Integration

apiVersion: v1
kind: ConfigMap
metadata:
  name: k8sgpt-webhook-config
data:
  WEBHOOK_URL: "https://your-service.com/webhook"
  WEBHOOK_SECRET: "your-secret"
  WEBHOOK_EVENTS: "analysis.completed,issue.detected"

• Use appropriate authentication for all integrations
• Implement rate limiting to prevent overloading
• Set up proper error handling and retries
• Monitor integration performance
• Keep integration configurations in version control
• Regularly test integrations to ensure they work