Implementing Kubernetes Network Policy With Calico

使用 Calico 实施 Kubernetes 网络策略

概述

Calico 是一个开源 CNI 插件，为 Kubernetes 集群提供细粒度网络策略执行能力。它实现了完整的 Kubernetes NetworkPolicy API，并通过 Calico 专属的 GlobalNetworkPolicy 进行了扩展，支持策略排序、拒绝规则和基于服务账户的选择器。

前提条件

• Kubernetes 集群（v1.24+）
• 已安装 Calico CNI（v3.26+）
• kubectl 和 calicoctl CLI 工具
• 集群管理员 RBAC 权限

安装 Calico

基于 Operator 的安装（推荐）

# 安装 Tigera Operator
kubectl create -f https://raw.githubusercontent.com/projectcalico/calico/v3.27.0/manifests/tigera-operator.yaml

# 安装 Calico 自定义资源
kubectl create -f https://raw.githubusercontent.com/projectcalico/calico/v3.27.0/manifests/custom-resources.yaml

# 验证安装
kubectl get pods -n calico-system
watch kubectl get pods -n calico-system

# 安装 calicoctl
kubectl apply -f https://raw.githubusercontent.com/projectcalico/calico/v3.27.0/manifests/calicoctl.yaml

验证 Calico 运行状态

# 检查 Calico Pod
kubectl get pods -n calico-system

# 检查 Calico 节点状态
kubectl exec -n calico-system calicoctl -- calicoctl node status

# 检查 IP 池
kubectl exec -n calico-system calicoctl -- calicoctl get ippool -o wide

Kubernetes NetworkPolicy

默认拒绝所有流量

# deny-all-ingress.yaml
apiVersion: networking.k8s.io/v1
kind: NetworkPolicy
metadata:
  name: default-deny-ingress
  namespace: production
spec:
  podSelector: {}
  policyTypes:
    - Ingress

---
# deny-all-egress.yaml
apiVersion: networking.k8s.io/v1
kind: NetworkPolicy
metadata:
  name: default-deny-egress
  namespace: production
spec:
  podSelector: {}
  policyTypes:
    - Egress

允许特定 Pod 间通信

# allow-frontend-to-backend.yaml
apiVersion: networking.k8s.io/v1
kind: NetworkPolicy
metadata:
  name: allow-frontend-to-backend
  namespace: production
spec:
  podSelector:
    matchLabels:
      app: backend
  policyTypes:
    - Ingress
  ingress:
    - from:
        - podSelector:
            matchLabels:
              app: frontend
      ports:
        - protocol: TCP
          port: 8080

允许 DNS 出口流量

# allow-dns-egress.yaml
apiVersion: networking.k8s.io/v1
kind: NetworkPolicy
metadata:
  name: allow-dns-egress
  namespace: production
spec:
  podSelector: {}
  policyTypes:
    - Egress
  egress:
    - to:
        - namespaceSelector: {}
      ports:
        - protocol: UDP
          port: 53
        - protocol: TCP
          port: 53

命名空间隔离

# allow-same-namespace.yaml
apiVersion: networking.k8s.io/v1
kind: NetworkPolicy
metadata:
  name: allow-same-namespace
  namespace: production
spec:
  podSelector: {}
  policyTypes:
    - Ingress
  ingress:
    - from:
        - podSelector: {}

Calico 专属策略

GlobalNetworkPolicy（集群级别）

# global-deny-external.yaml
apiVersion: projectcalico.org/v3
kind: GlobalNetworkPolicy
metadata:
  name: deny-external-ingress
spec:
  order: 100
  selector: "projectcalico.org/namespace != 'ingress-nginx'"
  types:
    - Ingress
  ingress:
    - action: Deny
      source:
        nets:
          - 0.0.0.0/0
      destination: {}

带拒绝规则的 Calico NetworkPolicy

# calico-deny-policy.yaml
apiVersion: projectcalico.org/v3
kind: NetworkPolicy
metadata:
  name: deny-database-from-frontend
  namespace: production
spec:
  order: 10
  selector: app == 'database'
  types:
    - Ingress
  ingress:
    - action: Deny
      source:
        selector: app == 'frontend'
    - action: Allow
      source:
        selector: app == 'backend'
      destination:
        ports:
          - 5432

基于服务账户的策略

# sa-based-policy.yaml
apiVersion: projectcalico.org/v3
kind: NetworkPolicy
metadata:
  name: allow-by-service-account
  namespace: production
spec:
  selector: app == 'api'
  ingress:
    - action: Allow
      source:
        serviceAccounts:
          names:
            - frontend-sa
            - monitoring-sa
  egress:
    - action: Allow
      destination:
        serviceAccounts:
          names:
            - database-sa

主机端点保护

# host-endpoint-policy.yaml
apiVersion: projectcalico.org/v3
kind: GlobalNetworkPolicy
metadata:
  name: restrict-host-ssh
spec:
  order: 10
  selector: "has(kubernetes.io/hostname)"
  applyOnForward: false
  types:
    - Ingress
  ingress:
    - action: Allow
      protocol: TCP
      source:
        nets:
          - 10.0.0.0/8
      destination:
        ports:
          - 22
    - action: Deny
      protocol: TCP
      destination:
        ports:
          - 22

Calico 策略层级

# security-tier.yaml
apiVersion: projectcalico.org/v3
kind: Tier
metadata:
  name: security
spec:
  order: 100

---
# platform-tier.yaml
apiVersion: projectcalico.org/v3
kind: Tier
metadata:
  name: platform
spec:
  order: 200

监控与故障排查

# 列出所有网络策略
kubectl get networkpolicy --all-namespaces

# 列出 Calico 专属策略
kubectl exec -n calico-system calicoctl -- calicoctl get networkpolicy --all-namespaces -o wide
kubectl exec -n calico-system calicoctl -- calicoctl get globalnetworkpolicy -o wide

# 检查特定端点的策略评估
kubectl exec -n calico-system calicoctl -- calicoctl get workloadendpoint -n production -o yaml

# 查看 Calico 日志
kubectl logs -n calico-system -l k8s-app=calico-node --tail=100

# 测试连通性
kubectl exec -n production frontend-pod -- wget -qO- --timeout=2 http://backend-svc:8080/health

最佳实践

1. 从默认拒绝开始 - 对每个命名空间应用拒绝所有策略，再允许特定流量
2. 统一使用标签 - 为 app、tier、environment 定义标签规范
3. 排序策略 - 使用 Calico 策略排序（order 字段）控制评估优先级
4. 先允许 DNS - 在应用出口拒绝策略前，始终先创建 DNS 出口规则
5. 使用 GlobalNetworkPolicy 建立集群级安全基线
6. 在预发布环境测试策略 - 应用策略后验证网络连通性
7. 监控被拒绝的流量 - 启用 Calico 流量日志以了解被阻断的连接
8. 使用层级 - 将策略组织为安全、平台和应用层级