resources:
  requests:
    cpu: "200m"      # 保证 0.2 核
    memory: "256Mi"  # 保证 256M 内存
  limits:
    cpu: "500m"      # 最多用 0.5 核
    memory: "512Mi"  # 最多用 512M 内存

# 1. 创建不同资源规格的 Pod
cat <<EOF | kubectl apply -f -
apiVersion: v1
kind: Pod
metadata:
  name: guaranteed-pod
spec:
  containers:
  - name: app
    image: nginx:alpine
    resources:
      requests:
        cpu: "100m"
        memory: "128Mi"
      limits:
        cpu: "100m"
        memory: "128Mi"
---
apiVersion: v1
kind: Pod
metadata:
  name: burstable-pod
spec:
  containers:
  - name: app
    image: nginx:alpine
    resources:
      requests:
        cpu: "100m"
        memory: "128Mi"
      limits:
        cpu: "500m"
        memory: "512Mi"
---
apiVersion: v1
kind: Pod
metadata:
  name: beste-ffort-pod
spec:
  containers:
  - name: app
    image: nginx:alpine
EOF

# 2. 查看 QoS 等级
kubectl get pod guaranteed-pod -o jsonpath='{.status.qosClass}'
echo
# Guaranteed

kubectl get pod burstable-pod -o jsonpath='{.status.qosClass}'
echo
# Burstable

kubectl get pod beste-ffort-pod -o jsonpath='{.status.qosClass}'
echo
# BestEffort

# 3. 模拟内存超限（OOMKilled）
cat <<EOF | kubectl apply -f -
apiVersion: v1
kind: Pod
metadata:
  name: oom-pod
spec:
  containers:
  - name: mem-eater
    image: busybox:1.36
    command:
    - sh
    - -c
    - |
      # 分配超过 limit 的内存
      dd if=/dev/zero of=/dev/shm/bigfile bs=100M count=10
      sleep 3600
    resources:
      limits:
        memory: "50Mi"
EOF

kubectl get pod oom-pod -w
# 观察 RESTARTS 递增，Last State: OOMKilled

kubectl describe pod oom-pod | grep -A5 "Last State"

# 4. 清理
kubectl delete pod guaranteed-pod burstable-pod beste-ffort-pod oom-pod

# 创建命名空间
kubectl create ns resource-lab

# 创建 LimitRange
cat <<EOF | kubectl apply -f -
apiVersion: v1
kind: LimitRange
metadata:
  name: default-limits
  namespace: resource-lab
spec:
  limits:
  - type: Container
    default:
      cpu: "200m"
      memory: "256Mi"
    defaultRequest:
      cpu: "100m"
      memory: "128Mi"
    max:
      cpu: "1"
      memory: "1Gi"
    min:
      cpu: "50m"
      memory: "64Mi"
EOF

# 创建不设 resources 的 Pod（自动应用默认值）
kubectl run auto-pod --image=nginx:alpine -n resource-lab

# 验证自动注入
kubectl get pod auto-pod -n resource-lab -o yaml | grep -A8 resources
# 应有默认的 requests 和 limits

# 尝试创建超过 max 的 Pod（应被拒绝）
cat <<EOF | kubectl apply -f -
apiVersion: v1
kind: Pod
metadata:
  name: over-limit-pod
  namespace: resource-lab
spec:
  containers:
  - name: app
    image: nginx:alpine
    resources:
      limits:
        cpu: "2"          # 超过 max.cpu=1
        memory: "256Mi"
EOF
# Error: [cpu: Invalid value: "2": must be less than or equal to cpu limit]

# 清理
kubectl delete pod auto-pod -n resource-lab

# 创建 ResourceQuota
cat <<EOF | kubectl apply -f -
apiVersion: v1
kind: ResourceQuota
metadata:
  name: team-quota
  namespace: resource-lab
spec:
  hard:
    requests.cpu: "2"
    requests.memory: "2Gi"
    limits.cpu: "4"
    limits.memory: "4Gi"
    pods: "10"
    persistentvolumeclaims: "5"
    services: "5"
EOF

# 查看配额
kubectl describe quota team-quota -n resource-lab

# 创建 Deployment 测试配额消耗
kubectl create deploy quota-test -n resource-lab --image=nginx:alpine --replicas=5

# 查看配额使用情况
kubectl describe quota team-quota -n resource-lab
# 可以看到 Used vs Hard

# 清理
kubectl delete deploy quota-test -n resource-lab

# 0. 确保 metrics-server 已安装
kubectl get pods -n kube-system -l k8s-app=metrics-server
# 如果没有，先安装：
kubectl apply -f https://github.com/kubernetes-sigs/metrics-server/releases/latest/download/components.yaml

# 1. 创建 Deployment（带资源请求）
cat <<EOF | kubectl apply -f -
apiVersion: apps/v1
kind: Deployment
metadata:
  name: hpa-demo
spec:
  replicas: 1
  selector:
    matchLabels:
      app: hpa-demo
  template:
    metadata:
      labels:
        app: hpa-demo
    spec:
      containers:
      - name: web
        image: nginx:alpine
        resources:
          requests:
            cpu: "50m"
            memory: "64Mi"
          limits:
            cpu: "200m"
            memory: "128Mi"
EOF

kubectl expose deploy hpa-demo --port=80

# 2. 创建 HPA
kubectl autoscale deployment hpa-demo --cpu=50% --min=1 --max=5

# 3. 查看 HPA
kubectl get hpa
# NAME       REFERENCE             TARGETS   MINPODS   MAXPODS   REPLICAS
# hpa-demo   Deployment/hpa-demo   0%/50%    1         5         1

kubectl describe hpa hpa-demo

# 4. 生成负载
kubectl run load-generator --image=busybox:1.36 --rm -it --restart=Never -- \
  sh -c 'while true; do wget -q -O- http://hpa-demo; done'

# 在另一个终端观察 HPA
kubectl get hpa hpa-demo -w
# 观察 TARGETS 上升，REPLICAS 自动增加

# 5. Ctrl+C 停止负载，观察缩容
kubectl get hpa hpa-demo -w
# REPLICAS 逐渐减少回 1

# 6. 清理
kubectl delete hpa hpa-demo
kubectl delete deploy hpa-demo
kubectl delete svc hpa-demo

# 排查清单
# 1. metrics-server 是否运行？
kubectl get pods -n kube-system -l k8s-app=metrics-server

# 2. 能否获取到指标？
kubectl top pods
kubectl top nodes

# 3. Deployment 是否设置了 resources.requests？
kubectl get deploy <name> -o yaml | grep -A5 resources
# HPA 基于百分比需要 requests 做基准

# 4. HPA 状态
kubectl describe hpa <name>
# 看 Events 区域的错误信息

# 查看被驱逐的 Pod
kubectl get pods --field-selector=status.phase=Failed

# 查看驱逐原因
kubectl describe pod <evicted-pod>
# The node was low on resource: memory.