Kubernetes吊舱无法到达

Question

我遇到了一个奇怪的问题，其中一个环境中的Pod不允许通过指定的端口进行请求

从另一个吊舱：

/app # wget 10.12.2.22:6000
Connecting to 10.12.2.22:6000 (10.12.2.22:6000)
wget: error getting response`

通过服务：

/app # wget services:6000
Connecting to services:6000 (10.15.249.82:6000)
wget: error getting response: Resource temporarily unavailable

从内部开始：

/app # wget localhost:6000
converted 'http://localhost:6000' (ANSI_X3.4-1968) -> 'http://localhost:6000' (UTF-8)
--2018-10-03 01:26:39--  http://localhost:6000/
Resolving localhost (localhost)... ::1, 127.0.0.1
Connecting to localhost (localhost)|::1|:6000... connected.
HTTP request sent, awaiting response... 404 Not Found
2018-10-03 01:26:39 ERROR 404: Not Found.

这是我的YAML：

---
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
  name: services
  namespace: production
  labels:
    name: services
spec:
  replicas: 1
  selector:
    matchLabels:
      app: services
      version: production
  template:
    metadata:
      name: services
      labels:
        app: services
        version: production
    spec:
      containers:
      - name: services
        image: gcr.io/myproject/services:v1.6.0
        imagePullPolicy: Always
        volumeMounts:
        - name: "gc-serviceaccount"
          readOnly: true
          mountPath: "/app/secrets"
        securityContext:
          privileged: true
        ports:
        - name: http
          containerPort: 3000
          protocol: TCP
        - name: sockets
          containerPort: 6000
          protocol: TCP
        env:
        - name: NODE_ENV
          value: "local"
        - name: PORTAL_HOST
          value: "portal.myproject.com"
        - name: PORTAL_HOSTNAME
          value: "http://portal"
        - name: DESIGNER_HOSTNAME
          value: "http://designer"
        - name: FILE_HOST
          value: "https://files.myproject.com"
        - name: STORAGE_BUCKET
          value: "production"
        - name: DATA_PATH
          value: ""
        - name: MONGO_DB_NAME
          value: "MyProject"
        - name: MONGO_REPLICA_SET
          value: "myproject"
        - name: PORT
          value: "3000"
        - name: REDIS_HOSTNAME
          value: "redis"
        - name: REDIS_PORT
          value: "6379"
        - name: SOCKETCLUSTER_WS_ENGINE
          value: "ws"
        - name: SOCKETCLUSTER_PORT
          value: "6000"
        - name: CLIENT_HOST
          value: "myproject.com"
        readinessProbe:
          httpGet:
            path: /healthz
            port: 3000
          periodSeconds: 1
          timeoutSeconds: 1
          successThreshold: 1
          failureThreshold: 10

      volumes:
      - name: "gc-serviceaccount"
        secret:
          secretName: "gc-serviceaccount"
---
apiVersion: v1
kind: Service
metadata:
  name: services
  namespace: production
spec:
  type: NodePort
  ports:
  - name: http
    port: 80
    targetPort: http
  - name: sockets
    port: 6000
    targetPort: sockets
  selector:
    app: services

值得注意的是，当尝试从端口3000上的另一个Pod连接到Pod时，我没有任何问题。

到目前为止，我已经创建了一个新环境并部署到该环境，但是我收到了相同的错误。

奇怪的是，我的其他较旧的环境运行良好，所以我想知道这是否是Kubernetes的问题？