linux运维、架构之路-K8s中部署Jenkins集群高可用

一、在K8s中 部署Jenkins优点

1、传统Jenkins集群架构一些问题

• Master发生故障时，整个流程都不可用
• Slave集群的环境配置不一样，来完成不同语言的编译打包，但是这些差异化的配置导致管理起来不方便，维护麻烦
• 资源分配不均衡，有的slave要运行的job出现排队等待，而有的salve处于空闲状态
• 资源浪费，每台slave可能是物理机或者虚拟机，当slave处于空闲状态时，也不能完全释放掉资源

2、K8s中Jenkins集群架构优点

• 当Jenkins Master接受到Build请求后，会根据配置的Label动态创建一个运行在Pod中的Jenkins Slave并注册到Master上，当运行完Job后，这个Slave会被注销并且这个Pod也会自动删除，恢复到最初的状态(这个策略可以设置)
• 服务高可用，当Jenkins Master出现故障时，Kubernetes会自动创建一个新的Jenkins Master容器，并且将Volume分配给新创建的容器，保证数据不丢失，从而达到集群服务高可用的作用
• 动态伸缩，合理使用资源，每次运行Job时，会自动创建一个Jenkins Slave，Job完成后，Slave自动注销并删除容器，资源自动释放，并且Kubernetes会根据每个资源的使用情况，动态分配slave到空闲的节点上创建，降低出现因某节点资源利用率高，降低出现因某节点利用率高出现排队的情况
• 扩展性好，当Kubernetes集群的资源严重不足导致Job排队等待时，可以很容器的添加一个Kubernetes Node到集群，从而实现扩展

二、K8s 部署Jenkins

1、集群环境

 2、k8s-node1部署NFS服务

①NFS服务端配置

mkdir -p /data/k8s
chown -R nfsnobody.nfsnobody /data
echo "/data/k8s 192.168.56.0/24(rw,no_root_squash,sync)" >/etc/exports
systemctl enable rpcbind
systemctl enable nfs
systemctl start rpcbind
systemctl start nf

②客户端挂载

systemctl start rpcbind
systemctl enable rpcbind
mkdir /data/k8s -p
mount -t nfs 192.168.56.11:/data/k8s /data/k8

3、创建Jenkins集群所需的YAML文件

①创建命名空间和存放Jenkins的YAML目录

kubectl create namespace demon
mkdir -p /opt/jenkins

②为Jenkins数据持久化存储创建一个PV

cat >/opt/jenkins/jenkins_pv.yaml <<EOF
apiVersion: v1
kind: PersistentVolume
metadata:
  name: opspv
spec:
  capacity:
    storage: 10Gi
  accessModes:
    - ReadWriteMany
  persistentVolumeReclaimPolicy: Delete
  nfs:
    server: 192.168.56.11
    path: /data/k8s

---
kind: PersistentVolumeClaim
apiVersion: v1
metadata:
  name: opspvc
  namespace: demon
spec:
  accessModes:
    - ReadWriteMany
  resources:
    requests:
      storage: 10Gi
EOF

③创建Jenkins集群权限serviceAccount文件

cat >/opt/jenkins/jenkins_rbac.yaml <<EOF
apiVersion: v1
kind: ServiceAccount
metadata:
  name: jenkins
  namespace: demon

---

kind: ClusterRole
apiVersion: rbac.authorization.k8s.io/v1beta1
metadata:
  name: jenkins
rules:
  - apiGroups: ["extensions", "apps"]
    resources: ["deployments"]
    verbs: ["create", "delete", "get", "list", "watch", "patch", "update"]
  - apiGroups: [""]
    resources: ["services"]
    verbs: ["create", "delete", "get", "list", "watch", "patch", "update"]
  - apiGroups: [""]
    resources: ["pods"]
    verbs: ["create","delete","get","list","patch","update","watch"]
  - apiGroups: [""]
    resources: ["pods/exec"]
    verbs: ["create","delete","get","list","patch","update","watch"]
  - apiGroups: [""]
    resources: ["pods/log"]
    verbs: ["get","list","watch"]
  - apiGroups: [""]
    resources: ["secrets"]
    verbs: ["get"]

---
apiVersion: rbac.authorization.k8s.io/v1beta1
kind: ClusterRoleBinding
metadata:
  name: jenkins
  namespace: demon
roleRef:
  apiGroup: rbac.authorization.k8s.io
  kind: ClusterRole
  name: jenkins
subjects:
  - kind: ServiceAccount
    name: jenkins
    namespace: demon
EOF

④创建Jenkins Deployment

cat jenkins_deployment.yaml 
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
  name: jenkins
  namespace: demon
spec:
  template:
    metadata:
      labels:
        app: jenkins
    spec:
      terminationGracePeriodSeconds: 10
      serviceAccount: jenkins
      containers:
      - name: jenkins
        image: jenkins/jenkins:lts
        imagePullPolicy: IfNotPresent
        ports:
        - containerPort: 8080                #外部访问端口
          name: web
          protocol: TCP
        - containerPort: 50000              #jenkins save发现端口
          name: agent
          protocol: TCP
        resources:
          limits:
            cpu: 1000m
            memory: 1Gi
          requests:
            cpu: 500m
            memory: 512Mi
        livenessProbe:
          httpGet:
            path: /login
            port: 8080
          initialDelaySeconds: 60          #容器初始化完成后，等待60秒进行探针检查
          timeoutSeconds: 5
          failureThreshold: 12          #当Pod成功启动且检查失败时，Kubernetes将在放弃之前尝试failureThreshold次。放弃生存检查意味着重新启动Pod。而放弃就绪检查，Pod将被标记为未就绪。默认为3.最小值为1
        readinessProbe:
          httpGet:
            path: /login
            port: 8080
          initialDelaySeconds: 60
          timeoutSeconds: 5
          failureThreshold: 12
        volumeMounts:                       #需要将jenkins_home目录挂载出来
        - name: jenkinshome
          subPath: jenkins
          mountPath: /var/jenkins_home
        env:
        - name: LIMITS_MEMORY
          valueFrom:
            resourceFieldRef:
              resource: limits.memory
              divisor: 1Mi
        - name: JAVA_OPTS
          value: -Xmx$(LIMITS_MEMORY)m -XshowSettings:vm -Dhudson.slaves.NodeProvisioner.initialDelay=0 -Dhudson.slaves.NodeProvisioner.MARGIN=50 -Dhudson.slaves.NodeProvisioner.MARGIN0=0.85 -Duser.timezone=Asia/Shanghai
      securityContext:
        fsGroup: 1000
      volumes:
      - name: jenkinshome
        persistentVolumeClaim:
          claimName: opspvc

⑤创建Jenkins SVC

cat >/opt/jenkins/jenkins_svc.yaml <<EOF
apiVersion: v1
kind: Service
metadata:
  name: jenkins
  namespace: demon
  labels:
    app: jenkins
spec:
  selector:
    app: jenkins
  type: NodePort
  ports:
  - name: web
    port: 8080
    targetPort: web
    nodePort: 30002
  - name: agent
    port: 50000
    targetPort: agent
EOF

⑥依次创建

[root@k8s-node1 jenkins]# ll
总用量 16
-rw-r--r-- 1 root root 2281 12月 27 14:59 jenkins_deployment.yaml
-rw-r--r-- 1 root root  404 12月 27 14:57 jenkins_pv.yaml
-rw-r--r-- 1 root root 1108 12月 27 14:58 jenkins_rbac.yaml
-rw-r--r-- 1 root root  285 12月 27 14:59 jenkins_svc.yaml

[root@k8s-node1 jenkins]# kubectl apply -f jenkins_pv.yaml 
persistentvolume/opspv created
persistentvolumeclaim/opspvc created

[root@k8s-node1 jenkins]# kubectl apply -f jenkins_rbac.yaml 
serviceaccount/jenkins created
clusterrole.rbac.authorization.k8s.io/jenkins created
clusterrolebinding.rbac.authorization.k8s.io/jenkins created

[root@k8s-node1 jenkins]# kubectl apply -f jenkins_deployment.yaml 
deployment.extensions/jenkins created

[root@k8s-node1 jenkins]# kubectl apply -f jenkins_svc.yaml 
service/jenkins create

⑦查看结果

[root@k8s-node1 jenkins]# kubectl get pv,pvc,pod,svc -n demon
NAME                     CAPACITY   ACCESS MODES   RECLAIM POLICY   STATUS    CLAIM          STORAGECLASS   REASON    AGE
persistentvolume/opspv   10Gi       RWX            Delete           Bound     demon/opspvc                            1h

NAME                           STATUS    VOLUME    CAPACITY   ACCESS MODES   STORAGECLASS   AGE
persistentvolumeclaim/opspvc   Bound     opspv     10Gi       RWX                           1h

NAME                           READY     STATUS    RESTARTS   AGE
pod/jenkins-6d7bc49b74-d9jxc   1/1       Running   0          1h

NAME              TYPE       CLUSTER-IP     EXTERNAL-IP   PORT(S)                          AGE
service/jenkins   NodePort   10.1.148.201   <none>        8080:30002/TCP,50000:26723/TCP   1h

8080：端口为我们jenkins访问端口  50000：端口为jenkins save发现端口

⑧通过浏览器访问集群任意IP的svc端口

管理员密码路径：持久化在/data/k8s下，所以jenkins的所有配置都在这下面

cat /data/k8s/jenkins/secrets/initialAdminPassword

 直接推荐安装即可

⑨安装完成后我们进入jenkins主页面

 ⑩enkins–>插件–>安装插件Kubernetes

 三、Jenkins中配置k8s

1、系统管理->系统配置

 2、配置拉到最下面找到Kubernetes插件

Name            配置的名称
Kubernetes URL  这里的URL是K8s内部的URL，实际上就是svcname  https://kubernetes.default.svc.cluster.local
Kubernetes Namespace k8s的命名空间 （实际上就是Jenkins所在的命名空间）

 3、Jenkins URL配置

Jenkins URL   这里的URL是jenkins的svc名称加上命名空间，实际上就是在k8s集群内部访问jenkins的一个方法，这里也不需要修改
http://jenkins.demon.svc.cluster.local:8080

 4、配置添 Jenkins Slave Pod模板

Name = Pod 名称 Namespave = Pod命名空间 Labels = Pod标签

5、容器的模板配置

 

 6、创建volume的配置

 Jenkins Master收到Build请求时，会根据配置的Label动态创建一个运行在Pod中的Jenkins Slave并注册到Master上，当Job运行完，这个Slave会被注销并且这个Pod也会自动删除，恢复到最初状态

四、测试验证

1、新建Job选择流水线

 2、流水线Pipeline

def label = "jenkins-slave"
podTemplate(label: label, cloud: 'kubernetes')
{
node(label) {
        stage('pull code') {
            echo "拉取代码"
        }
        stage('build') {
            echo "代码编译"
        }
        stage('SonarQube') {
            echo "质量扫描"
        }
    }
}

3、执行效果