• Helm 部署 skywalkingv8.4.0



    1. 调用链简介

      在分布式架构、微服务以及 k8s 生态相关技术环境下,对应用的请求链路进行追踪(也叫做 APM,Application Performance Management)是非常有必要的,链路追踪简单来说就是将应用从流量到达前端开始,一直到最后端的数据库核心,中间经过的每一层请求链路的完整行为都记录下来,而且通过可视化的形式实现链路信息查询、依赖关系、性能分析、拓扑展示等等,利用链路追踪系统可以很好的帮我们定位问题,这是常规监控手段实现起来比较困难的。

    常见商业版本:

    • 听云
    • 博睿宏远

    常见开源版本:

    • Skywalking:中国,个人开源,目前隶属于 Apache 基金会,作者近期刚刚入选 Apache 首位中国董事
    • Pinpoint:韩国,个人开源
    • Zipkin:美国,Twitter 公司开源
    • Cat:中国,美团开源

    2. 环境

    • K8S v1.22.5 集群
    主机 IP
    master 192.168.10.100
    node01 192.168.10.101
    node02 192.168.10.102
    • Elasticsearch v7.12.0
    • Skywalking
    skywalking-oap-server:后端服务
    skywalking-ui:ui 前端
    skywalking-es-init:初始化 es 集群数据使用
    elasticsearch:存储 skywalking 的数据指标
    

    本次编写时候:

    skywalking 最高版本 9.1.0

    elasticsearch 最高版本 8.3.2

    3. K8S 集群部署 nfs 环境

    3.1 创建命名空间

    [root@master ~]# kubectl create ns efk
    namespace/efk created
    [root@master ~]# kubectl get ns
    NAME              STATUS   AGE
    default           Active   22h
    efk               Active   2s
    ingress-nginx     Active   22h
    istio-system      Active   19h
    kube-node-lease   Active   22h
    kube-public       Active   22h
    kube-system       Active   22h
    metallb-system    Active   22h
    

    3.2 创建 NFS

    ### 这里就将 nfs-server 安装在 master 节点
    
    # 安装 nfs-utils、rpcbind 软件包(===所有节点===)
    yum -y install nfs-utils rpcbind
    
    # 创建目录
    sudo mkdir -p /nfsdata
     
    # 添加权限
    sudo chmod 777 -R /nfsdata
     
    # 编辑文件,添加以下内容
    sudo vim /etc/exports
    /nfsdata 192.168.10.0/24(rw,no_root_squash,sync)
    
    # 重启服务
    systemctl start rpcbind && systemctl enable rpcbind
    systemctl start nfs && systemctl enable nfs(所有节点)
    
    # 配置生效
    exportfs -rv
    
    # 查看共享目录
    sudo showmount -e 192.168.10.100
    # 返回值如下,表示创建成功
    Export list for 192.168.10.100:
    /nfsdata	192.168.108.*
    

    3.3 创建 StorageClass

    external-storage/nfs-client/deploy at master · kubernetes-retired/external-storage (github.com)

    rbac.yaml:创建 serviceaccount

    apiVersion: v1
    kind: ServiceAccount
    metadata:
      name: nfs-client-provisioner
      namespace: default 
    ---
    kind: ClusterRole
    apiVersion: rbac.authorization.k8s.io/v1
    metadata:
      name: nfs-client-provisioner-runner
    rules:
      - apiGroups: [""]
        resources: ["persistentvolumes"]
        verbs: ["get", "list", "watch", "create", "delete"]
      - apiGroups: [""]
        resources: ["persistentvolumeclaims"]
        verbs: ["get", "list", "watch", "update"]
      - apiGroups: ["storage.k8s.io"]
        resources: ["storageclasses"]
        verbs: ["get", "list", "watch"]
      - apiGroups: [""]
        resources: ["events"]
        verbs: ["create", "update", "patch"]
    ---
    kind: ClusterRoleBinding
    apiVersion: rbac.authorization.k8s.io/v1
    metadata:
      name: run-nfs-client-provisioner
    subjects:
      - kind: ServiceAccount
        name: nfs-client-provisioner
        namespace: default
    roleRef:
      kind: ClusterRole
      name: nfs-client-provisioner-runner
      apiGroup: rbac.authorization.k8s.io
    ---
    kind: Role
    apiVersion: rbac.authorization.k8s.io/v1
    metadata:
      name: leader-locking-nfs-client-provisioner
      namespace: default
    rules:
      - apiGroups: [""]
        resources: ["endpoints"]
        verbs: ["get", "list", "watch", "create", "update", "patch"]
    ---
    kind: RoleBinding
    apiVersion: rbac.authorization.k8s.io/v1
    metadata:
      name: leader-locking-nfs-client-provisioner
    subjects:
      - kind: ServiceAccount
        name: nfs-client-provisioner
        namespace: default
    roleRef:
      kind: Role
      name: leader-locking-nfs-client-provisioner
      apiGroup: rbac.authorization.k8s.io
    

    prvisor-deployment.yaml:创建 nfs-client-provisioner,要和 rbac 一个 ns

    apiVersion: apps/v1
    kind: Deployment
    metadata:
      name: nfs-client-provisioner
      labels:
        app: nfs-client-provisioner
      namespace: default 
    spec:
      replicas: 1
      selector:
        matchLabels:
          app: nfs-client-provisioner
      strategy:
        type: Recreate
      selector:
        matchLabels:
          app: nfs-client-provisioner
      template:
        metadata:
          labels:
            app: nfs-client-provisioner
        spec:
          serviceAccountName: nfs-client-provisioner
          containers:
            - name: nfs-client-provisioner
              image: quay.io/external_storage/nfs-client-provisioner:latest
              volumeMounts:
                - name: nfs-client-root
                  mountPath: /persistentvolumes
              env:
                - name: PROVISIONER_NAME
                  value: wuchang-nfs-storage 
                - name: NFS_SERVER
                  value: 192.168.10.100   #NFS Server IP地址
                - name: NFS_PATH 
                  value: /nfsdata        #NFS挂载卷
          volumes:
            - name: nfs-client-root
              nfs:
                server: 192.168.10.100    #NFS Server IP地址
                path: /nfsdata           #NFS 挂载卷
    

    storageclass.yaml:创建 storageclass

    apiVersion: storage.k8s.io/v1
    kind: StorageClass
    metadata:
      name: managed-nfs-storage
    provisioner: wuchang-nfs-storage
    
    # 允许 pvc 创建后扩容
    allowVolumeExpansion: True
    
    parameters:
    # 资源删除策略,如果为 "true" 则表示删除 PVC 时,同时删除绑定的 PV
      archiveOnDelete: "false"
    

    按顺序执行

    kubectl apply -f rbac.yaml
    kubectl get serviceaccount
    
    kubectl apply -f prvisor-deployment.yaml
    kubectl get deploy
    
    kubectl apply -f storageclass.yaml
    kubectl get sc
    

    4. K8S 安装 ES

    可事先下载镜像:

    docker pull elasticsearch:x.x.x
    docker pull apache/skywalking-oap-server:9.0.0
    docker pull apache/skywalking-ui:9.0.0
    

    es-pvc.yaml:serviceaccount 和 pv-deploy 要放在 default 下,pvc 可以指定任意 ns。

    apiVersion: v1
    kind: PersistentVolumeClaim
    metadata:
      name: es-pvc
      namespace: default
    spec:
      storageClassName: "managed-nfs-storage" #指定动态 PV 名称
      accessModes:
        - ReadWriteMany
      resources:
        requests:
          storage: 10Gi
    

    es-svc.yaml

    apiVersion: v1
    kind: Service
    metadata:
      name: elasticsearch-single
      namespace: default
    spec:
      ports:
      - port: 9200
        protocol: TCP
        targetPort: 9200
      selector:
        k8s-app: elasticsearch-single
    

    elasticsearch-single.yaml

    apiVersion: apps/v1
    kind: Deployment
    metadata:
      name: elasticsearch-single
      namespace: default
      labels:
        k8s-app: elasticsearch-single
    spec:
      replicas: 1
      selector:
        matchLabels:
          k8s-app: elasticsearch-single
      template:
        metadata:
          labels:
            k8s-app: elasticsearch-single
        spec:
          containers:
          - image: elasticsearch:7.12.0		# 目前最高 8.3.2
            name: elasticsearch-single
            resources:
              limits:
                cpu: 2
                memory: 3Gi
              requests:
                cpu: 0.5 
                memory: 500Mi
            env:
              - name: "discovery.type"
                value: "single-node"
              - name: ES_JAVA_OPTS
                value: "-Xms512m -Xmx2g" 
            ports:
            - containerPort: 9200
              name: db
              protocol: TCP
            volumeMounts:
            - name: elasticsearch-data
              mountPath: /usr/share/elasticsearch/data
          volumes:
          - name: elasticsearch-data
            persistentVolumeClaim:
              claimName: es-pvc
    

    执行

    kubectl apply -f es-pvc.yaml
    kubectl get pv,pvc
    
    kubectl apply -f es-svc.yaml
    kubectl get svc
    
    kubectl apply -f elasticsearch-single.yaml
    kubectl get pod
    

    故障描述:

    PVC 显示创建不成功:kubectl get pvc -n efk 显示 Pending,这是由于版本太高导致的。k8sv1.20 以上版本默认禁止使用 selfLink。(selfLink:通过 API 访问资源自身的 URL,例如一个 Pod 的 link 可能是 /api/v1/namespaces/ns36aa8455/pods/sc-cluster-test-1-6bc58d44d6-r8hld)。

    故障解决:

    [root@k8sm storage]# vi /etc/kubernetes/manifests/kube-apiserver.yaml
    apiVersion: v1
    ···
        - --tls-private-key-file=/etc/kubernetes/pki/apiserver.key
        - --feature-gates=RemoveSelfLink=false # 添加这个配置
    重启下kube-apiserver.yaml
    
    # 如果是二进制安装的 k8s,执行 systemctl restart kube-apiserver
    # 如果是 kubeadm 安装的 k8s
    [root@k8sm manifests]# ps aux|grep kube-apiserver
    [root@k8sm manifests]# kill -9 [Pid]	# 有可能自动重启
    [root@k8sm manifests]# kubectl apply -f /etc/kubernetes/manifests/kube-apiserver.yaml
    ...
    [root@master ~]# kubectl get pods -A | grep kube-apiserver-master	# 查看时间
    ......
    [root@k8sm storage]# kubectl get pvc	# 查看 pvc 显示 Bound
    NAME     STATUS   VOLUME                                     CAPACITY   ACCESS MODES   STORAGECLASS   AGE
    my-pvc   Bound    pvc-ae9f6d4b-fc4c-4e19-8854-7bfa259a3a04   1Gi        RWX            example-nfs    13m
    

    5. 安装 skywalking

    5.1 安装 Helm

    curl https://raw.githubusercontent.com/helm/helm/master/scripts/get-helm-3 | bash
    curl http://49.232.8.65/shell/helm/helm-v3.5.0_install.sh | bash
    ------------------------------------------------------------------------------------
    [root@master ~]# curl https://raw.githubusercontent.com/helm/helm/master/scripts/get-helm-3 | bash
      % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                     Dload  Upload   Total   Spent    Left  Speed
    100 11156  100 11156    0     0  17974      0 --:--:-- --:--:-- --:--:-- 17964
    Downloading https://get.helm.sh/helm-v3.9.0-linux-amd64.tar.gz
    Verifying checksum... Done.
    Preparing to install helm into /usr/local/bin
    helm installed into /usr/local/bin/helm
    

    5.2 初始化 skywalking 的 charts 配置

    clone helm 仓库

    git clone https://github.com/apache/skywalking-kubernetes
    cd skywalking-kubernetes/chart && ls
    

    添加 ES repo:即使使用外部 ES 也要添加这个 repo,否则会导致依赖错误

    helm repo add elastic https://helm.elastic.co
    helm dep up skywalking
    export SKYWALKING_RELEASE_NAME=skywalking
    export SKYWALKING_RELEASE_NAMESPACE=skywalking
    -------------------------------------------------------
    [root@master ~/skywalking-kubernetes/chart]# ls
    skywalking
    [root@master ~/skywalking-kubernetes/chart]# helm dep up skywalking
    Hang tight while we grab the latest from your chart repositories...
    ...Successfully got an update from the "elastic" chart repository
    Update Complete. ⎈Happy Helming!⎈
    Saving 1 charts
    Downloading elasticsearch from repo https://helm.elastic.co/
    Deleting outdated charts
    [root@master ~/skywalking-kubernetes/chart]# export SKYWALKING_RELEASE_NAME=skywalking
    [root@master ~/skywalking-kubernetes/chart]# export SKYWALKING_RELEASE_NAMESPACE=skywalking
    

    创建 skywalking 的 namespace

    [root@master ~]# kubectl create namespace skywalking
    namespace/skywalking created
    [root@master ~]# kubectl get ns
    NAME              STATUS   AGE
    default           Active   2d1h
    efk               Active   124m
    ingress-nginx     Active   2d
    istio-system      Active   46h
    kube-node-lease   Active   2d1h
    kube-public       Active   2d1h
    kube-system       Active   2d1h
    metallb-system    Active   2d
    skywalking        Active   4s
    

    5.3 配置 skywalking 的 vaules 配置参数

    初始化完成后需要自行调整配置文件:

    1️⃣ 配置 oap-server 使用外部 ES,values-my-es-o1.yaml(自己创建)

    2️⃣ 使用 values 自带的 es 的配置示例 values-my-es.yaml

    预先下载镜像文件:
    docker pull skywalking.docker.scarf.sh/apache/skywalking-oap-server:9.1.0
    docker pull skywalking.docker.scarf.sh/apache/skywalking-ui:9.1.0

    修改 values.yaml

    [root@master ~/skywalking-kubernetes/chart/skywalking]# vim values.yaml
    ----------------------------------------
    ......
    image:
        repository: skywalking.docker.scarf.sh/apache/skywalking-oap-server
      改为:
        repository: docker.mirrors.ustc.edu.cn/apache/skywalking-oap-server
    ......
    

    如果使用外部 es

    # skywalking 目前最高 9.1.0
    [root@master ~/skywalking-kubernetes/chart/skywalking]# cat values-my-es-01.yaml
    oap:
      image:
        tag: 8.4.0-es7 
      storageType: elasticsearch7
    
    ui:
      image:
        tag: 8.4.0
      service:
        type: NodePort
        externalPort: 80
        internalPort: 8080
        nodePort: 30008
    elasticsearch:
      enabled: false
      config: 
        # {SERVICE_NAME}.{NAMESPACE_NAME}.svc.cluster.local
        host: elasticsearch-single.default  # elasticsearch-single 是服务名,default 是命名空间
        port:
          http: 9200
        # user: "elastic"         # [optional]
        # password: "elastic"     # [optional]
        # es 没有使用账号密码
    

    5.4 helm 安装 skywalking 8.4.0

    cd /root/skywalking-kubernetes/chart/
    
    --- 方法一:直接指定版本安装,不适用外部 es
    helm install "${SKYWALKING_RELEASE_NAME}" skywalking -n "${SKYWALKING_RELEASE_NAMESPACE}"   --set oap.image.tag=8.4.0-es7   --set oap.storageType=elasticsearch7   --set ui.image.tag=8.4.0   --set elasticsearch.imageTag=7.12.0
    
    --- 方法二:使用外部 es 命令(设置环境变量有点多此一举)
    helm install "${SKYWALKING_RELEASE_NAME}" skywalking -n "${SKYWALKING_RELEASE_NAMESPACE}" -f ./skywalking/values-my-es-01.yaml
    
    -----------------------------------------------------------------------------------------
    [root@master ~/skywalking-kubernetes/chart]# helm install skywalking skywalking -n  skywalking  -f ./skywalking/values-my-es-01.yaml
    NAME: skywalking
    LAST DEPLOYED: Sat Jul  9 15:05:23 2022
    NAMESPACE: skywalking
    STATUS: deployed
    REVISION: 1
    TEST SUITE: None
    NOTES:
    ************************************************************************
    *                                                                      *
    *                 SkyWalking Helm Chart by SkyWalking Team             *
    *                                                                      *
    ************************************************************************
    
    Thank you for installing skywalking.
    
    Your release is named skywalking.
    
    Learn more, please visit https://skywalking.apache.org/
    
    Get the UI URL by running these commands:
      export NODE_PORT=$(kubectl get --namespace skywalking -o jsonpath="{.spec.ports[0].nodePort}" services skywalking-ui)
      export NODE_IP=$(kubectl get nodes --namespace skywalking -o jsonpath="{.items[0].status.addresses[0].address}")
      echo http://$NODE_IP:$NODE_PORT
    

    卸载 skywalking

    helm uninstall skywalking -n skywalking

    持续查看 pod 安装进度

    [root@master ~/skywalking-kubernetes/chart]# kubectl get pod -n skywalking -w
    [root@master ~/skywalking-kubernetes/chart]# kubectl get pod -n skywalking -w
    NAME                              READY   STATUS            RESTARTS   AGE
    skywalking-es-init--1-fnnbn       0/1     PodInitializing   0          42s
    skywalking-oap-7596f94959-97ccs   0/1     PodInitializing   0          42s
    skywalking-oap-7596f94959-t5sxv   0/1     PodInitializing   0          42s
    skywalking-ui-7957d9fb5f-drpwg    1/1     Running           0          42s
    ......
    

    临时对外暴露 skywalking 端口,我用了 NodePort 的方法开放了端口,生产中也可以使用 ingress 的方式开放

    export POD_NAME=$(kubectl get pods --namespace skywalking -l "app=skywalking,release=skywalking,component=ui" -o jsonpath="{.items[0].metadata.name}")
    kubectl port-forward $POD_NAME 8080:8080 --namespace skywalking
    

    查看 skywalking 的访问 URL:k8s master/node ip + nodeport

    export NODE_PORT=$(kubectl get --namespace skywalking -o jsonpath="{.spec.ports[0].nodePort}" services skywalking-ui)
    export NODE_IP=$(kubectl get nodes --namespace skywalking -o jsonpath="{.items[0].status.addresses[0].address}")
    echo http://$NODE_IP:$NODE_PORT
    
    -----------------------------------------------------------------------------------------
    [root@master ~]#kubectl get --namespace skywalking -o jsonpath="{.spec.ports[0].nodePort}" services skywalking-ui
    30008
    [root@master ~]#kubectl get nodes --namespace skywalking -o jsonpath="{.items[0].status.addresses[0].address}"
    192.168.10.65
    

    运行状态检查

    # docker.mirrors.ustc.edu.cn/apache/skywalking-oap-server   8.4.0-es7  # 这个镜像最好提前下载
    # 显示 PodInitializing 就是因为镜像一直没下载下来
    [root@master ~]#kubectl get pods,svc -o wide -n skywalking
    NAME                                        READY   STATUS            RESTARTS      AGE   IP            NODE     NOMINATED NODE   READINESS GATES
    pod/elasticsearch-single-6768b6454b-l5tds   1/1     Running           0             31m   10.244.2.17   node02   <none>           <none>
    pod/skywalking-es-init--1-kqkz2             0/1     Completed         0             28m   10.244.1.32   node01   <none>           <none>
    pod/skywalking-oap-666d7ffb45-bd2bl         0/1     PodInitializing   0             28m   10.244.2.18   node02   <none>           <none>
    pod/skywalking-oap-666d7ffb45-kpl7p         1/1     Running           1 (24m ago)   28m   10.244.1.31   node01   <none>           <none>
    pod/skywalking-ui-9c4c5f495-vwnrz           1/1     Running           0             28m   10.244.1.30   node01   <none>           <none>
    
    NAME                           TYPE        CLUSTER-IP       EXTERNAL-IP   PORT(S)               AGE   SELECTOR
    service/elasticsearch-single   ClusterIP   10.102.62.187    <none>        9200/TCP              31m   k8s-app=elasticsearch-single
    service/skywalking-oap         ClusterIP   10.101.90.191    <none>        11800/TCP,12800/TCP   28m   app=skywalking,component=oap,release=skywalking
    service/skywalking-ui          NodePort    10.111.254.170   <none>        80:30008/TCP          28m   app=skywalking,component=ui,release=skywalking
    

    有时候 init 和 oap pod 运行不成功,显示等待 es container 创建,这是 oap 没有检测到 es 存储,可能是对接出了问题,有点玄学,需要排查。

    访问

    http://nodeIP:nodePort


    K8s 环境 nfs 动态存储卷部署

    Helm 部署 skywalking

    k8s 跨 namespace 访问服务

    helm v3 在 k8s 上面的部署 skywalking

    k8s 部署 elasticsearch(包含数据挂载VOLUME)

    k8s 安装 elasticsearch 集群

    公博义

    K3S 集群安装 helmv3 与使用

    Skywalking 实战

    【Docker】之部署 skywalking 实现全链路监控功能

    Kubernetes 中部署 ES 集群及运维 --- 华为云 重点


  • 相关阅读:
    LoadRunner 技巧之 集合点设置
    LoadRunner 技巧之 IP欺骗 (推荐)
    JMeter 聚合报告之 90% Line 参数说明
    python基础学习笔记(二)
    LoadRunner 技巧之HTML 与 URL两种录制模式分析
    LoadRunner 技巧之 检查点
    LoadRunner 技巧之 思考时间设置
    一个JSlider的例子
    JApplet添加图片
    java.util.Arrays.asList 的小问题
  • 原文地址:https://www.cnblogs.com/shenyuanhaojie/p/16464416.html
Copyright © 2020-2023  润新知