kubernetes入門到實戰（十）守護和任務型控制器—（01）

時間  2020-07-09
標籤 kubernetes 入門 實戰 守護 任務 控制器 欄目 Java 简体版
原文   原文鏈接

1. DaemonSet控制器

1.1 DaemonSet簡介

介紹DaemonSet時咱們先來思考一個問題：相信你們都接觸過監控系統好比zabbix，監控系統須要在被監控機安裝一個agent，安裝agent一般會涉及到如下幾個場景：node

  • 全部節點都必須安裝agent以便採集監控數據
  • 新加入的節點須要配置agent，手動或者運行腳本
  • 節點下線後須要手動在監控系統中刪除

kubernetes中常常涉及到在node上安裝部署應用，它是如何解決上述的問題的呢？答案是DaemonSet。DaemonSet守護進程簡稱DS，適用於在全部節點或部分節點運行一個daemon守護進程，如監控咱們安裝部署時網絡插件kube-flannel和kube-proxy，DaemonSet具備以下特色：linux

  • DaemonSet確保全部節點運行一個Pod副本
  • 指定節點運行一個Pod副本，經過標籤選擇器或者節點親和性
  • 新增節點會自動在節點增長一個Pod
  • 移除節點時垃圾回收機制會自動清理Pod

DaemnonSet控制器

DaemonSet適用於每一個node節點均須要部署一個守護進程的場景，常見的場景例如：web

  • 日誌採集agent，如fluentd或logstash
  • 監控採集agent，如Prometheus Node Exporter,Sysdig Agent,Ganglia gmond
  • 分佈式集羣組件，如Ceph MON，Ceph OSD，glusterd，Hadoop Yarn NodeManager等
  • k8s必要運行組件，如網絡flannel，weave，calico，kube-proxy等

安裝k8s時默認在kube-system命名空間已經安裝了有兩個DaemonSet，分別爲kube-flannel-ds-amd64和kube-proxy，分別負責flannel overlay網絡的互通和service代理的實現，能夠經過以下命令查看：算法

  1. 查看kube-system命令空間的DaemonSet列表，當前集羣有三個node節點，因此每一個DS會運行三個Pod副本
[root@node-1 ~]# kubectl get ds -n kube-system 
NAME                    DESIRED   CURRENT   READY   UP-TO-DATE   AVAILABLE   NODE SELECTOR                   AGE
kube-flannel-ds-amd64   3         3         3       3            3           beta.kubernetes.io/arch=amd64   46d
kube-proxy              3         3         3       3            3           beta.kubernetes.io/os=linux     46d
  1. 查看Pod的副本狀況，能夠看到DaemonSet在每一個節點都運行一個Pod

kubernetes入門到實戰（十）守護和任務型控制器—（01）

1.2 DaemonSet定義

DaemonSet的定義和Deployment定義使用相相似，須要定義apiVersion，Kind，metadata和spec屬性信息，spec中不須要定義replicas個數，spec.template即定義DS生成容器的模版信息，以下是運行一個fluentd-elasticsearch鏡像容器的daemon守護進程，運行在每一個node上經過fluentd採集日誌上報到ElasticSearch。docker

  1. 經過yaml文件定義DaemonSet
[root@node-1 happycloudlab]# cat fluentd-es-daemonset.yaml 
apiVersion: apps/v1              #api版本信息
kind: DaemonSet                  #類型爲DaemonSet
metadata:                        #元數據信息
  name: fluentd-elasticsearch
  namespace: kube-system        #運行的命名空間
  labels:
    k8s-app: fluentd-logging
spec:                          #DS模版
  selector:
    matchLabels:
      name: fluentd-elasticsearch
  template:
    metadata:
      labels:
        name: fluentd-elasticsearch
    spec:
      tolerations:
      - key: node-role.kubernetes.io/master
        effect: NoSchedule
      containers:            #容器信息
      - name: fluentd-elasticsearch
        image: quay.io/fluentd_elasticsearch/fluentd:v2.5.2
        resources:          #resource資源
          limits:
            memory: 200Mi
          requests:
            cpu: 100m
            memory: 200Mi
        volumeMounts:      #掛載存儲，agent須要到這些目錄採集日誌
        - name: varlog
          mountPath: /var/log
        - name: varlibdockercontainers
          mountPath: /var/lib/docker/containers
          readOnly: true
      terminationGracePeriodSeconds: 30
      volumes:            #將主機的目錄以hostPath的形式掛載到容器Pod中。
      - name: varlog
        hostPath:
          path: /var/log
      - name: varlibdockercontainers
        hostPath:
          path: /var/lib/docker/containers

DaemonSet定義注意事項：api

  • daemonset.spec.template定義Pod的模板信息，包含的metadata信息須要和selector保持一致
  • template必須定義RestartPolicy的策略，切策略值爲Always，保障服務異常時能自動重啓恢復
  • Pod運行在特定節點，支持指定調度策略，如nodeSelector，Node affinity，實現靈活調度
  1. 生成DaemonSet
[root@node-1 happycloudlab]# kubectl apply -f fluentd-es-daemonset.yaml 
daemonset.apps/fluentd-elasticsearch created
  1. 查看DaemonSet列表
[root@node-1 happycloudlab]# kubectl get daemonsets -n kube-system  fluentd-elasticsearch 
NAME                    DESIRED   CURRENT   READY   UP-TO-DATE   AVAILABLE   NODE SELECTOR   AGE
fluentd-elasticsearch   3         3         3       3            3           <none>          16s
  1. 查看node上運行Pod的狀況,在NODE列能夠看到每一個node都運行了一個Pod
[root@node-1 happycloudlab]# kubectl get pods -n kube-system -o wide |grep fluentd 
fluentd-elasticsearch-blpqb      1/1     Running   0          3m7s   10.244.2.79      node-3   <none>           <none>
fluentd-elasticsearch-ksdlt      1/1     Running   0          3m7s   10.244.0.11      node-1   <none>           <none>
fluentd-elasticsearch-shtkh      1/1     Running   0          3m7s   10.244.1.64      node-2   <none>           <none>
  1. 查看DaemonSet詳情，能夠看到DaemonSet支持RollingUpdate滾動更新策略
[root@node-1 happycloudlab]# kubectl get daemonsets -n kube-system fluentd-elasticsearch -o yaml
apiVersion: extensions/v1beta1
kind: DaemonSet
metadata:
  annotations:
    kubectl.kubernetes.io/last-applied-configuration: |
      {"apiVersion":"apps/v1","kind":"DaemonSet","metadata":{"annotations":{},"labels":{"k8s-app":"fluentd-logging"},"name":"fluentd-elasticsearch","namespace":"kube-system"},"spec":{"selector":{"matchLabels":{"name":"fluentd-elasticsearch"}},"template":{"metadata":{"labels":{"name":"fluentd-elasticsearch"}},"spec":{"containers":[{"image":"quay.io/fluentd_elasticsearch/fluentd:v2.5.2","name":"fluentd-elasticsearch","resources":{"limits":{"memory":"200Mi"},"requests":{"cpu":"100m","memory":"200Mi"}},"volumeMounts":[{"mountPath":"/var/log","name":"varlog"},{"mountPath":"/var/lib/docker/containers","name":"varlibdockercontainers","readOnly":true}]}],"terminationGracePeriodSeconds":30,"tolerations":[{"effect":"NoSchedule","key":"node-role.kubernetes.io/master"}],"volumes":[{"hostPath":{"path":"/var/log"},"name":"varlog"},{"hostPath":{"path":"/var/lib/docker/containers"},"name":"varlibdockercontainers"}]}}}}
  creationTimestamp: "2019-10-30T15:19:20Z"
  generation: 1
  labels:
    k8s-app: fluentd-logging
  name: fluentd-elasticsearch
  namespace: kube-system
  resourceVersion: "6046222"
  selfLink: /apis/extensions/v1beta1/namespaces/kube-system/daemonsets/fluentd-elasticsearch
  uid: c2c02c48-9f93-48f3-9d6c-32bfa671db0e
spec:
  revisionHistoryLimit: 10
  selector:
    matchLabels:
      name: fluentd-elasticsearch
  template:
    metadata:
      creationTimestamp: null
      labels:
        name: fluentd-elasticsearch
    spec:
      containers:
      - image: quay.io/fluentd_elasticsearch/fluentd:v2.5.2
        imagePullPolicy: IfNotPresent
        name: fluentd-elasticsearch
        resources:
          limits:
            memory: 200Mi
          requests:
            cpu: 100m
            memory: 200Mi
        terminationMessagePath: /dev/termination-log
        terminationMessagePolicy: File
        volumeMounts:
        - mountPath: /var/log
          name: varlog
        - mountPath: /var/lib/docker/containers
          name: varlibdockercontainers
          readOnly: true
      dnsPolicy: ClusterFirst
      restartPolicy: Always             #重啓策略必須爲Always，保障異常時能自動恢復
      schedulerName: default-scheduler  #默認調度策略
      securityContext: {}
      terminationGracePeriodSeconds: 30
      tolerations:
      - effect: NoSchedule
        key: node-role.kubernetes.io/master
      volumes:
      - hostPath:
          path: /var/log
          type: ""
        name: varlog
      - hostPath:
          path: /var/lib/docker/containers
          type: ""
        name: varlibdockercontainers
  templateGeneration: 1
  updateStrategy:  #滾動更新策略
    rollingUpdate:
      maxUnavailable: 1
    type: RollingUpdate
status:
  currentNumberScheduled: 3
  desiredNumberScheduled: 3
  numberAvailable: 3
  numberMisscheduled: 0
  numberReady: 3
  observedGeneration: 1
  updatedNumberScheduled: 3

1.3 滾動更新與回滾

  1. 更新鏡像至最新版本
[root@node-1 ~]# kubectl set image daemonsets fluentd-elasticsearch fluentd-elasticsearch=quay.io/fluentd_elasticsearch/fluentd:latest -n kube-system
daemonset.extensions/fluentd-elasticsearch image updated
  1. 查看滾動更新狀態
[root@node-1 ~]# kubectl rollout status daemonset -n kube-system fluentd-elasticsearch 
Waiting for daemon set "fluentd-elasticsearch" rollout to finish: 1 out of 3 new pods have been updated...
Waiting for daemon set "fluentd-elasticsearch" rollout to finish: 1 out of 3 new pods have been updated...
Waiting for daemon set "fluentd-elasticsearch" rollout to finish: 1 out of 3 new pods have been updated...
Waiting for daemon set "fluentd-elasticsearch" rollout to finish: 2 out of 3 new pods have been updated...
Waiting for daemon set "fluentd-elasticsearch" rollout to finish: 2 out of 3 new pods have been updated...
Waiting for daemon set "fluentd-elasticsearch" rollout to finish: 2 out of 3 new pods have been updated...
Waiting for daemon set "fluentd-elasticsearch" rollout to finish: 2 of 3 updated pods are available...
daemon set "fluentd-elasticsearch" successfully rolled out
  1. 查看DaemonSet詳情，能夠看到DS滾動更新的過程：DaemonSet先將node上的pod刪除而後再建立

DaemonSet滾動更新過程

  1. 查看DaemonSet滾動更新版本，REVSION 1爲初始的版本
[root@node-1 ~]# kubectl rollout history daemonset -n kube-system fluentd-elasticsearch 
daemonset.extensions/fluentd-elasticsearch 
REVISION  CHANGE-CAUSE
1         <none>
2         <none>
  1. 更新回退，若是配置沒有符合到預期能夠回滾到原始的版本
[root@node-1 ~]# kubectl rollout undo daemonset -n kube-system fluentd-elasticsearch --to-revision=1
daemonset.extensions/fluentd-elasticsearch rolled back
  1. 確認版本回退狀況

DaemonSet版本回退

  1. 觀察版本回退的過程，回退的過程和和滾動更新過程相似，先刪除Pod再建立

DaemonSet回退過程

  1. 刪除DaemonSet
[root@node-1 ~]# kubectl delete daemonsets -n kube-system fluentd-elasticsearch 
daemonset.extensions "fluentd-elasticsearch" deleted
[root@node-1 ~]# kubectl get pods -n kube-system |grep fluentd
fluentd-elasticsearch-d6f6f      0/1     Terminating   0          110m

1.4 DaemonSet調度

DaemonSet經過kubernetes默認的調度器scheduler會在全部的node節點上運行一個Pod副本，能夠經過以下三種方式將Pod運行在部分節點上：網絡

  • 指定nodeName節點運行
  • 經過標籤運行nodeSelector
  • 經過親和力調度node Affinity和node Anti-affinity

DaemonSet調度算法用於實現將Pod運行在特定的node節點上，以下以經過node affinity親和力將Pod調度到部分的節點上node-2上爲例。app

  1. 爲node添加一個app=web的labels
[root@node-1 happycloudlab]# kubectl get nodes --show-labels 
NAME     STATUS   ROLES    AGE   VERSION   LABELS
node-1   Ready    master   47d   v1.15.3   beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=node-1,kubernetes.io/os=linux,node-role.kubernetes.io/master=
node-2   Ready    <none>   47d   v1.15.3   app=web,beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=node-2,kubernetes.io/os=linux
node-3   Ready    <none>   47d   v1.15.3   beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=node-3,kubernetes.io/os=linux
  1. 添加node affinity親和力調度算法，requiredDuringSchedulingIgnoredDuringExecution設置基本須要知足條件，preferredDuringSchedulingIgnoredDuringExecution設置優選知足條件
[root@node-1 happycloudlab]# cat fluentd-es-daemonset.yaml 
apiVersion: apps/v1
kind: DaemonSet
metadata:
  name: fluentd-elasticsearch
  namespace: kube-system
  labels:
    k8s-app: fluentd-logging
spec:
  selector:
    matchLabels:
      name: fluentd-elasticsearch
  template:
    metadata:
      labels:
        name: fluentd-elasticsearch
    spec:
      tolerations:
      - key: node-role.kubernetes.io/master
        effect: NoSchedule
      containers:
      - name: fluentd-elasticsearch
        image: quay.io/fluentd_elasticsearch/fluentd:v2.5.2
        resources:
          limits:
            memory: 200Mi
          requests:
            cpu: 100m
            memory: 200Mi
        volumeMounts:
        - name: varlog
          mountPath: /var/log
        - name: varlibdockercontainers
          mountPath: /var/lib/docker/containers
          readOnly: true
      affinity:
        nodeAffinity:
          preferredDuringSchedulingIgnoredDuringExecution:  #優先知足條件
          - weight: 1
            preference:
              matchExpressions:
              - key: app 
                operator: In
                values:
                - web 
          requiredDuringSchedulingIgnoredDuringExecution:  #要求知足條件
            nodeSelectorTerms:
            - matchExpressions:
              - key: kubernetes.io/hostname
                operator: In
                values:
                - node-2
                - node-3
      terminationGracePeriodSeconds: 30
      volumes:
      - name: varlog
        hostPath:
          path: /var/log
      - name: varlibdockercontainers
        hostPath:
          path: /var/lib/docker/containers
  1. 生成DS，並查看列表
[root@node-1 happycloudlab]# kubectl delete ds -n kube-system fluentd-elasticsearch 
daemonset.extensions "fluentd-elasticsearch" deleted

[root@node-1 happylau]# kubectl get daemonsets -n kube-system fluentd-elasticsearch 
NAME                    DESIRED   CURRENT   READY   UP-TO-DATE   AVAILABLE   NODE SELECTOR   AGE
fluentd-elasticsearch   1         1         1       1            1           <none>          112s
  1. 校驗Pod運行的狀況，DaemonSet的Pod調度到node-2節點上
[root@node-1 happycloudlab]# kubectl get pods -n kube-system -o wide 
NAME                             READY   STATUS    RESTARTS   AGE     IP               NODE     NOMINATED NODE   READINESS GATES          <none>
fluentd-elasticsearch-9kngs      1/1     Running   0          2m39s   10.244.1.82      node-2   <none>           <none>

寫在最後

本文介紹了kubernetes中DaemonSet控制器，DS控制器能確保全部的節點運行一個特定的daemon守護進程，此外經過nodeSelector或node Affinity可以實現將Pod調度到特定的node節點。elasticsearch

參考文檔

DaemonSet：https://kubernetes.io/docs/concepts/workloads/controllers/daemonset/分佈式

相關文章
相關標籤/搜索
每日一句
    每一个你不满意的现在，都有一个你没有努力的曾经。
本站公眾號
   歡迎關注本站公眾號,獲取更多信息
聯繫我們 最近搜索 最新文章 沪ICP备13005482号-10 MyBatis教程 SQL 教程 MySQL教程 Java 教程 Thymeleaf 教程 Hibernate教程 Spring教程 Redis教程