Шаблоны обнаружения сервисов в Kubernetes

$ kubectl get endpoints kubernetes-dashboard -n kubernetes-dashboard -o yaml
 
apiVersion: v1
kind: Endpoints
metadata:
  name: kubernetes-dashboard
  namespace: kubernetes-dashboard
subsets:
addresses:
  - ip: 10.244.0.23
    nodeName: worker-node1
    targetRef:
      kind: Pod
      name: kubernetes-dashboard-78c79f97b4-phxtd
  - ip: 10.244.1.31
    nodeName: worker-node2
    targetRef:
      kind: Pod
      name: kubernetes-dashboard-78c79f97b4-q2vkd
  # ... и ещё десятки или сотни записей при большом масштабе
  ports:
  - port: 8443
    protocol: TCP

# Масштабирование CoreDNS
kubectl scale --replicas=5 deployment/coredns -n kube-system

.:53 {
    errors
    health {
      lameduck 5s
    }
    ready
    kubernetes cluster.local in-addr.arpa ip6.arpa {
      pods insecure
      fallthrough in-addr.arpa ip6.arpa
    }
    cache 30 {
      success 9984 3600 # Увеличен размер кеша успешных запросов
      denial 9984 5     # Увеличен размер кеша негативных запросов
    }
    prometheus :9153
    forward . /etc/resolv.conf {
      max_concurrent 1000 # Увеличено количество одновременных запросов
    }
    loop
    reload
    loadbalance
}

# Примеры важных метрик CoreDNS
coredns_dns_request_count_total        # Общее количество запросов
coredns_dns_request_duration_seconds   # Время отклика
coredns_cache_hits_total               # Попадания в кеш
coredns_cache_misses_total             # Промахи мимо кеша

apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
  name: coredns
  namespace: kube-system
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: coredns
  minReplicas: 2
  maxReplicas: 10
  metrics:
  - type: Resource
    resource:
      name: cpu
      target:
        type: Utilization
        averageUtilization: 60
  - type: Resource
    resource:
      name: memory
      target:
        type: Utilization
        averageUtilization: 80

apiVersion: apps/v1
kind: DaemonSet
metadata:
  name: node-local-dns
  namespace: kube-system
spec:
  selector:
    matchLabels:
      k8s-app: node-local-dns
  template:
    metadata:
      labels:
        k8s-app: node-local-dns
    spec:
      containers:
      - name: node-cache
        image: k8s.gcr.io/dns/k8s-dns-node-cache:1.17.3
        resources:
          limits:
            memory: 170Mi
          requests:
            cpu: 100m
            memory: 70Mi
        args: [ "-localip", "169.254.20.10", "-conf", "/etc/coredns/Corefile" ]
        # ... остальная конфигурация

apiVersion: v1
kind: Service
metadata:
name: my-service
spec:
selector:
  app: my-app
ports:
port: 80
  targetPort: 8080

apiVersion: v1
kind: Service
metadata:
name: my-headless-service
spec:
clusterIP: None  # Это делает сервис "безголовым"
selector:
  app: my-stateful-app
ports:
port: 80
  targetPort: 8080

# Custom Resource для мультикластерного сервиса
apiVersion: multicluster.example.com/v1
kind: FederatedService
metadata:
name: global-payment-service
spec:
selector:
  service: payment-service
  clusters:
  - name: us-east
    namespace: prod
  - name: eu-west
    namespace: prod

readinessProbe:
httpGet:
  path: /health
  port: 8080
initialDelaySeconds: 5
periodSeconds: 10

apiVersion: monitoring.coreos.com/v1
kind: ServiceMonitor
metadata:
name: payment-service-monitor
spec:
selector:
  matchLabels:
    app: payment-service
endpoints:
port: metrics
  interval: 15s

spec:
strategy:
  type: RollingUpdate
  rollingUpdate:
    maxSurge: 1
    maxUnavailable: 0

lifecycle:
preStop:
  exec:
    command: ["sh", "-c", "sleep 10 && /app/shutdown.sh"]

topologyKeys:
"kubernetes.io/hostname"
"topology.kubernetes.io/zone"
"topology.kubernetes.io/region"
"*"

# Пример включения Istio-инжекции для namespace
apiVersion: v1
kind: Namespace
metadata:
  name: my-namespace
  labels:
    istio-injection: enabled

apiVersion: networking.istio.io/v1alpha3
kind: VirtualService
metadata:
  name: payment-service
spec:
  hosts:
  - payment-service
  http:
  - route:
    - destination:
        host: payment-service-v1
        subset: prod
      weight: 80
    - destination:
        host: payment-service-v2
        subset: canary
      weight: 20

# Headless Service для StatefulSet
apiVersion: v1
kind: Service
metadata:
  name: cassandra
spec:
  clusterIP: None  # The magic is here
  selector:
    app: cassandra
  ports:
  - port: 9042

apiVersion: v1
kind: Service
metadata:
  name: nginx
  annotations:
    external-dns.alpha.kubernetes.io/hostname: nginx.example.org
spec:
  type: LoadBalancer
  ports:
  - port: 80
    targetPort: 80
  selector:
    app: nginx

apiVersion: integrations.example.com/v1
kind: LegacyServiceRegistration
metadata:
name: payment-processor
spec:
serviceRef:
  name: payment-service
  namespace: financial
legacySystem:
  url: "https://legacy-registry.example.com"
  credentials:
    secretRef:
      name: legacy-credentials
  registrationPath: "/services/register"

# Пример Consul Service
apiVersion: consul.hashicorp.com/v1alpha1
kind: ServiceDefaults
metadata:
name: payment-service
spec:
protocol: "http"
---
apiVersion: consul.hashicorp.com/v1alpha1
kind: ServiceIntentions
metadata:
name: payment-service
spec:
destination:
  name: payment-service
sources:
name: web-frontend
  action: allow

apiVersion: cloud.example.com/v1
kind: GlobalService
metadata:
name: user-service
spec:
selector:
  service: user-service
visibility: global  # Может быть: global, regional или local
locations:
provider: aws
  region: us-east-1
provider: gcp
  region: europe-west1
loadBalancingPolicy: geo  # Может быть: geo, latency, failover
healthCheck:
  path: /health
  interval: 10s

# Пример настройки Anycast с помощью BGP на узле Kubernetes
gobgp global rib add 10.96.0.10/32 nexthop 192.168.1.10

apiVersion: networking.istio.io/v1alpha3
kind: VirtualService
metadata:
  name: reviews
spec:
  hosts:
  - reviews
  http:
  - match:
    - headers:
        end-user:
          exact: jason
    route:
    - destination:
        host: reviews
        subset: v2
  - route:
    - destination:
        host: reviews
        subset: v1
      weight: 75
    - destination:
        host: reviews
        subset: v3
      weight: 25

apiVersion: networking.istio.io/v1alpha3
kind: DestinationRule
metadata:
  name: reviews
spec:
  host: reviews
  trafficPolicy:
    outlierDetection:
      consecutiveErrors: 5
      interval: 30s
      baseEjectionTime: 1m
  subsets:
  - name: v1
    labels:
      version: v1
  - name: v2
    labels:
      version: v2
  - name: v3
    labels:
      version: v3

apiVersion: getambassador.io/v2
kind: Mapping
metadata:
  name: payment-service
spec:
  prefix: /payment/
  service: payment-service.default:8080
  timeout_ms: 5000
  retry_policy:
    retry_on: "5xx"
    num_retries: 3

apiVersion: getambassador.io/v2
kind: Mapping
metadata:
  name: payment-service-stable
spec:
  prefix: /payment/
  service: payment-service-stable.default:8080
  weight: 90
---
apiVersion: getambassador.io/v2
kind: Mapping
metadata:
  name: payment-service-canary
spec:
  prefix: /payment/
  service: payment-service-canary.default:8080
  weight: 10

apiVersion: consul.hashicorp.com/v1alpha1
kind: ServiceDefaults
metadata:
  name: payment-processor
spec:
  protocol: "http"
  mesh: true
  expose:
    paths:
      - path: /api/payments
        protocol: http
        local_path_port: 8080

service {
  name = "legacy-inventory"
  port = 8000
  connect {
    sidecar_service {}
  }
  checks = [
    {
      id = "http-check"
      http = "http://localhost:8000/health"
      interval = "10s"
      timeout = "1s"
    }
  ]
}

apiVersion: consul.hashicorp.com/v1alpha1
kind: ServiceIntentions
metadata:
  name: payment-to-inventory
spec:
  destination:
    name: legacy-inventory
  sources:
    - name: payment-processor
      action: allow

apiVersion: telemetry.istio.io/v1alpha1
kind: Telemetry
metadata:
  name: mesh-default
spec:
  tracing:
    - providers:
        - name: otel
      randomSamplingPercentage: 100.0
  metrics:
    - providers:
        - name: prometheus