HPA

# 간단한 HPA 생성 (CPU 50% 기준)
kubectl autoscale deployment web-app \
  --cpu-percent=50 \
  --min=2 \
  --max=10

# HPA 상태 확인
kubectl get hpa web-app
# NAME      REFERENCE            TARGETS   MINPODS   MAXPODS   REPLICAS
# web-app   Deployment/web-app   35%/50%   2         10        3

# HPA 상세 정보 확인
kubectl describe hpa web-app

# HPA 이벤트 및 스케일링 이력 확인
kubectl get hpa web-app -o yaml

# 실시간 모니터링
kubectl get hpa -w

# HPA 삭제
kubectl delete hpa web-app

# HPA v2 전체 설정 예제
apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
  name: web-app-hpa
  namespace: production
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: web-app
  minReplicas: 2
  maxReplicas: 20
  metrics:
  # CPU 기반 스케일링
  - type: Resource
    resource:
      name: cpu
      target:
        type: Utilization
        averageUtilization: 50
  # 메모리 기반 스케일링
  - type: Resource
    resource:
      name: memory
      target:
        type: Utilization
        averageUtilization: 70
  # 스케일 다운 안정화 설정
  behavior:
    scaleDown:
      stabilizationWindowSeconds: 300  # 5분
      policies:
      - type: Percent
        value: 10
        periodSeconds: 60  # 분당 최대 10% 감소
    scaleUp:
      stabilizationWindowSeconds: 0  # 즉시 스케일 업
      policies:
      - type: Percent
        value: 100
        periodSeconds: 15  # 15초마다 최대 2배 증가
      - type: Pods
        value: 4
        periodSeconds: 15  # 15초마다 최대 4개 증가

# 커스텀/외부 메트릭 기반 HPA
apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
  name: queue-processor-hpa
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: queue-processor
  minReplicas: 1
  maxReplicas: 50
  metrics:
  # Prometheus 커스텀 메트릭 (요청 처리량)
  - type: Pods
    pods:
      metric:
        name: http_requests_per_second
      target:
        type: AverageValue
        averageValue: "100"  # Pod당 100 RPS 유지
  # 외부 메트릭 (Kafka 큐 길이)
  - type: External
    external:
      metric:
        name: kafka_consumergroup_lag
        selector:
          matchLabels:
            topic: orders
            consumergroup: order-processor
      target:
        type: AverageValue
        averageValue: "1000"  # 큐 길이 1000 기준
---
# 필수: Deployment에 리소스 요청 설정
apiVersion: apps/v1
kind: Deployment
metadata:
  name: web-app
spec:
  template:
    spec:
      containers:
      - name: web
        resources:
          requests:
            cpu: "100m"    # HPA 계산에 필수
            memory: "256Mi"
          limits:
            cpu: "500m"
            memory: "512Mi"