K8s学习

cat > app.py << 'EOF'
from fastapi import FastAPI
import socket
import uvicorn

app = FastAPI()

@app.get("/")
def hello():
    hostname = socket.gethostname()
    return {"message": f"✅ 我的后端服务运行中！Pod 名称: {hostname}"}

@app.get("/health")
def health():
    return {"status": "healthy", "hostname": socket.gethostname()}

if __name__ == "__main__":
    uvicorn.run(app, host="0.0.0.0", port=5000)
EOF

cat > requirements.txt << 'EOF'
fastapi
uvicorn
EOF

创建DockerFile

cat > Dockerfile << 'EOF'
FROM python:3.11-slim

WORKDIR /app

# 复制依赖文件并安装
COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt

# 复制应用代码
COPY app.py .

# 暴露端口
EXPOSE 5000

# 使用 uvicorn 运行（性能更好）
CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "5000"]
EOF

# 构建镜像（使用 Docker）
docker build -t my-app:v1 .
# 查看镜像
docker images | grep my-app

导入镜像到 Containerd

# 因为你的集群使用 containerd，需要导入镜像
docker save my-app:v1 -o my-app.tar
sudo ctr -n k8s.io images import my-app.tar

# 验证导入成功
sudo crictl images | grep my-app

# 先复制 tar 文件到 node01
scp my-app.tar node01:~/

# 然后 SSH 到 node01 执行导入
ssh node01 "sudo ctr -n k8s.io images import ~/my-app.tar"

创建Pod，Deployment

cat > deployment.yaml << 'EOF'
apiVersion: apps/v1
kind: Deployment
metadata:
  name: my-app-deployment
spec:
  replicas: 3
  selector:
    matchLabels:
      app: my-app
  template:
    metadata:
      labels:
        app: my-app
    spec:
      containers:
      - name: my-app
        image: my-app:v1
        ports:
        - containerPort: 5000
        imagePullPolicy: IfNotPresent
EOF

# 创建 deployment
kubectl apply -f deployment.yaml

# 查看 deployment 状态
kubectl get deployments

# 查看 Pod 状态（应该很快变成 Running）
kubectl get pods -w

测试

# 方法1：端口转发到本地
kubectl port-forward deployment/my-backend 5000:5000

# 打开另一个终端，测试
curl http://localhost:5000/

# 方法2：查看 Pod 日志
kubectl logs -l app=my-backend

kubectl delete pod my-backend-xxxxxx-xxxxx

设置 Requests 和 Limits

cat > deployment.yaml << 'EOF'
apiVersion: apps/v1
kind: Deployment
metadata:
  name: my-app-deployment
spec:
  replicas: 2  # 初始 2 个副本
  selector:
    matchLabels:
      app: my-app
  template:
    metadata:
      labels:
        app: my-app
    spec:
      containers:
      - name: my-app
        image: my-app:v1
        ports:
        - containerPort: 5000
        imagePullPolicy: IfNotPresent
        resources:
          requests:
            cpu: "100m"      # 请求 0.1 核 CPU
            memory: "128Mi"  # 请求 128MB 内存
          limits:
            cpu: "200m"      # 最多使用 0.2 核 CPU
            memory: "256Mi"  # 最多使用 256MB 内存
EOF

# 应用更新
kubectl apply -f deployment.yaml

创建 HPA（Horizontal Pod Autoscaler）

cat > hpa.yaml << 'EOF'
apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
  name: my-app-hpa
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: my-app-deployment
  minReplicas: 2
  maxReplicas: 4
  metrics:
  - type: Resource
    resource:
      name: cpu
      target:
        type: Utilization
        averageUtilization: 50  # CPU 使用率超过 50% 时扩容
EOF

# 创建 HPA
kubectl apply -f hpa.yaml

# 查看 HPA 状态
kubectl get hpa

创建 Service

cat > service.yaml << 'EOF'
apiVersion: v1
kind: Service
metadata:
  name: my-app-service
spec:
  selector:
    app: my-app
  ports:
  - port: 5000
    targetPort: 5000
EOF

kubectl apply -f service.yaml

创建 HPA

cat > hpa.yaml << 'EOF'
apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
  name: my-app-hpa
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: my-app-deployment
  minReplicas: 2
  maxReplicas: 4
  metrics:
  - type: Resource
    resource:
      name: cpu
      target:
        type: Utilization
        averageUtilization: 50
EOF

kubectl apply -f hpa.yaml

# 查看 HPA 状态
kubectl get hpa

安装 metrics-server

# 下载并安装 metrics-server
kubectl apply -f https://github.com/kubernetes-sigs/metrics-
# 查看状态
kubectl get pods -n kube-system -l k8s-app=metrics-server

# 1. 删除所有 metrics-server 相关资源
kubectl delete deployment metrics-server -n kube-system --ignore-not-found=true
kubectl delete service metrics-server -n kube-system --ignore-not-found=true
kubectl delete apiservice v1beta1.metrics.k8s.io --ignore-not-found=true

# 2. 等待删除完成
sleep 5

# 3. 重新部署正确的配置
cat <<EOF | kubectl apply -f -
apiVersion: apps/v1
kind: Deployment
metadata:
  name: metrics-server
  namespace: kube-system
  labels:
    k8s-app: metrics-server
spec:
  replicas: 1
  selector:
    matchLabels:
      k8s-app: metrics-server
  template:
    metadata:
      labels:
        k8s-app: metrics-server
    spec:
      serviceAccountName: metrics-server
      containers:
      - name: metrics-server
        image: registry.k8s.io/metrics-server/metrics-server:v0.6.4
        args:
        - --kubelet-insecure-tls
        - --kubelet-preferred-address-types=InternalIP,Hostname
        - --cert-dir=/tmp
        ports:
        - containerPort: 10250
          name: https
          protocol: TCP
        volumeMounts:
        - name: tmp-dir
          mountPath: /tmp
      volumes:
      - name: tmp-dir
        emptyDir: {}
---
apiVersion: v1
kind: Service
metadata:
  name: metrics-server
  namespace: kube-system
spec:
  ports:
  - port: 443
    protocol: TCP
    targetPort: 10250
  selector:
    k8s-app: metrics-server
---
apiVersion: apiregistration.k8s.io/v1
kind: APIService
metadata:
  name: v1beta1.metrics.k8s.io
spec:
  service:
    name: metrics-server
    namespace: kube-system
  group: metrics.k8s.io
  version: v1beta1
  insecureSkipTLSVerify: true
  groupPriorityMinimum: 100
  versionPriority: 100
EOF

# 4. 等待 Pod 启动
sleep 10

# 5. 查看 Pod 状态（应该是 Running）
kubectl get pods -n kube-system -l k8s-app=metrics-server

# 6. 查看日志
kubectl logs -n kube-system -l k8s-app=metrics-server --tail=20

kubectl top nodes

创建压测 Pod

cat > load-generator.yaml << 'EOF'
apiVersion: v1
kind: Pod
metadata:
  name: load-generator
spec:
  containers:
  - name: loader
    image: busybox
    command:
    - /bin/sh
    - -c
    - |
      echo "开始压测..."
      while true; do
        wget -q -O- http://my-app-service:5000/ > /dev/null 2>&1
      done
EOF

kubectl apply -f load-generator.yaml

监控扩容

kubectl get pods -l app=my-app -w
kubectl get hpa -w
watch -n 2 'kubectl top pods -l app=my-app'

清理实验环境

# 删除压测 Pod
kubectl delete -f load-generator.yaml --ignore-not-found=true
kubectl delete pod load-generator-1 load-generator-2 load-generator-3 --ignore-not-found=true

# 等待5分钟，HPA 会自动缩容回 2
kubectl get pods -l app=my-app -w

# 完全清理（可选）
kubectl delete hpa my-app-hpa
kubectl delete service my-app-service
kubectl delete deployment my-app-deployment