Question

我在Jenkins的某些工作中看到了此错误

Cannot contact jenkins-slave-l65p0-0f7m0: hudson.remoting.ChannelClosedException: Channel "unknown": Remote call on JNLP4-connect connection from 100.99.111.187/100.99.111.187:46776 failed. The channel is closing down or has closed down

我有一个詹金斯大师-奴隶设置。

在从属服务器上找到以下日志

java.nio.channels.ClosedChannelException
    at org.jenkinsci.remoting.protocol.NetworkLayer.onRecvClosed(NetworkLayer.java:154)
    at org.jenkinsci.remoting.protocol.impl.NIONetworkLayer.ready(NIONetworkLayer.java:142)
    at org.jenkinsci.remoting.protocol.IOHub$OnReady.run(IOHub.java:795)
    at jenkins.util.ContextResettingExecutorService$1.run(ContextResettingExecutorService.java:28)
    at jenkins.security.ImpersonatingExecutorService$1.run(ImpersonatingExecutorService.java:59)
    at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
    at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
    at java.lang.Thread.run(Thread.java:748)

Jenkins位于kubernetes集群上。

apiVersion: apps/v1beta1
kind: StatefulSet
metadata:
  namespace: default
  name: jenkins-deployment
spec:
  serviceName: "jenkins-pod"
  replicas: 1
  template:
    metadata:
      labels:
        app: jenkins-pod
    spec:
      initContainers:
      - name: volume-mount-hack
        image: busybox
        command: ["sh", "-c", "chmod -R 777 /usr/mnt"]
        volumeMounts:
        - name: jenkinsdir
          mountPath: /usr/mnt
      containers:
      - name: jenkins-container
         imagePullPolicy: Always
        readinessProbe:
          exec:
            command:
              - curl
              - http://localhost:8080/login
              - -o
              - /dev/null
        livenessProbe:
          httpGet:
            path: /login
            port: 8080
          initialDelaySeconds: 120
          periodSeconds: 10
        env:
         - name: JAVA_OPTS
           value: "-Dhudson.slaves.NodeProvisioner.initialDelay=0 -Dhudson.slaves.NodeProvisioner.MARGIN=50 -Dhudson.slaves.NodeProvisioner.MARGIN0=0.85"
        resources:
          requests:
            memory: "7100Mi"
            cpu: "2000m"
        ports:
        - name: http-port
          containerPort: 8080
        - name: jnlp-port
          containerPort: 50000
        volumeMounts:
          - mountPath: /var/run
            name: docker-sock
          - mountPath: /var/jenkins_home
            name: jenkinsdir
      volumes:
        - name: jenkinsdir
          persistentVolumeClaim:
            claimName: "jenkins-persistence"
        - name: docker-sock
          hostPath:
            path: /var/run
---
apiVersion: v1
kind: Service
metadata:
  namespace: default
  name: jenkins
  labels:
    app: jenkins
spec:
  type: NodePort
  ports:
  - name: http
    port: 8080
    targetPort: 8080
    nodePort: 30099
    protocol: TCP
  selector:
    app: jenkins-pod
---
apiVersion: v1
kind: Service
metadata:
  namespace: default
  name: jenkins-external
  annotations:
        service.beta.kubernetes.io/aws-load-balancer-internal: 0.0.0.0/0
  labels:
    app: jenkins
spec:
  type: LoadBalancer
  ports:
  - name: http
    port: 8080
    targetPort: 8080
    protocol: TCP
  selector:
    app: jenkins-pod
---
apiVersion: policy/v1beta1
kind: PodDisruptionBudget
metadata:
  name: jenkins-master-pdb
  namespace: default
spec:
  maxUnavailable: 0
  selector:
    matchLabels:
      app: jenkins-pod
---
apiVersion: policy/v1beta1
kind: PodDisruptionBudget
metadata:
  name: jenkins-slave-pdb
  namespace: default
spec:
  maxUnavailable: 0
  selector:
    matchLabels:
      jenkins: slave
---
kind: Service
apiVersion: v1
metadata:
  name: jenkins-discovery
  namespace: default
  labels:
    app: jenkins
spec:
  selector:
    app: jenkins-pod
  ports:
    - protocol: TCP
      port: 50000
      targetPort: 50000
      name: slaves

我怀疑这与kubernetes有什么关系，但仍然没有解决。

Answer 1

我假设您正在使用Jenkins Kubernetes Plugin，

您可以在Timeout in seconds for Jenkins connection下增加Kubernetes Pod template。它可以解决您的问题。

Timeout in seconds for Jenkins connection的描述：

以秒为单位指定Jenkins应该等待JNLP的时间代理建立连接。值应为正整数，默认值为100。

Answer 2

您是否在Jenkins本身中配置了JNLP端口？它位于“管理Jenkins”>“配置全局安全性”>“代理”中。单击“固定”单选按钮（因为您已经分配了TCP端口）。将“ JNLP代理的TCP端口”设置为50000。

Answer 3

我认为“詹金斯-奴隶”不是一个有效的名称。您可以尝试将其重命名为“ jnlp”

在这里解释：

This was related to this issue. If the name of the custom agent is not jnlp, then another agent with the default jnlp image is created. This explains messages like channel already closed etc..

Jenkins从属JNLP4-连接超时

3 个答案: