Kubernetes 部署 Nebula 图数据库集群

2020 年 2 月 26 日
筆記

Kubernetes 是什么

Kubernetes 是一个开源的，用于管理云平台中多个主机上的容器化的应用，Kubernetes 的目标是让部署容器化的应用简单并且高效，Kubernetes 提供了应用部署，规划，更新，维护的一种机制。

Kubernetes 在设计结构上定义了一系列的构建模块，其目的是为了提供一个可以部署、维护和扩展应用程序的机制，组成 Kubernetes 的组件设计概念为松耦合和可扩展的，这样可以使之满足多种不同的工作负载。可扩展性在很大程度上由 Kubernetes API 提供，此 API 主要被作为扩展的内部组件以及 Kubernetes 上运行的容器来使用。

Kubernetes 主要由以下几个核心组件组成：

etcd 保存了整个集群的状态
apiserver 提供了资源操作的唯一入口，并提供认证、授权、访问控制、API注册和发现等机制
controller manager 负责维护集群的状态，比如故障检测、自动扩展、滚动更新等
scheduler 负责资源的调度，按照预定的调度策略将Pod调度到相应的机器上
kubelet 负责维护容器的生命周期，同时也负责 Volume和网络的管理
Container runtime 负责镜像管理以及 Pod 和容器的真正运行（CRI）
kube-proxy 负责为 Service 提供 cluster 内部的服务发现和负载均衡

除了核心组件，还有一些推荐的 Add-ons：

kube-dns 负责为整个集群提供 DNS 服务
Ingress Controller 为服务提供外网入口
Heapster 提供资源监控
Dashboard 提供 GUI
Federation 提供跨可用区的集群
Fluentd-elasticsearch 提供集群日志采集、存储与查询

Kubernetes 和数据库

数据库容器化是最近的一大热点，那么 Kubernetes 能为数据库带来什么好处呢？

故障恢复: Kubernetes 提供故障恢复的功能，数据库应用如果宕掉，Kubernetes 可以将其自动重启，或者将数据库实例迁移到集群中其他节点上
存储管理: Kubernetes 提供了丰富的存储接入方案，数据库应用能透明地使用不同类型的存储系统
负载均衡: Kubernetes Service 提供负载均衡功能，能将外部访问平摊给不同的数据库实例副本上
水平拓展: Kubernetes 可以根据当前数据库集群的资源利用率情况，缩放副本数目，从而提升资源的利用率

目前很多数据库，如：MySQL，MongoDB 和 TiDB 在 Kubernetes 集群中都能运行很良好。

Nebula Graph在Kubernetes中的实践

Nebula Graph 是一个分布式的开源图数据库，主要组件有：Query Engine 的 graphd，数据存储的 storaged，和元数据的 meted。在 Kubernetes 实践过程中，它主要给图数据库 Nebula Graph 带来了以下的好处：

Kubernetes 能分摊 nebula graphd，metad 和 storaged 不副本之间的负载。graphd，metad 和 storaged 可以通过 Kubernetes 的域名服务自动发现彼此。
通过 storageclass，pvc 和 pv 可以屏蔽底层存储细节，无论使用本地卷还是云盘，Kubernetes 均可以屏蔽这些细节。
通过 Kubernetes 可以在几秒内成功部署一套 Nebula 集群，Kubernetes 也可以无感知地实现 Nebula 集群的升级。
Nebula 集群通过 Kubernetes 可以做到自我恢复，单体副本 crash，Kubernetes 可以重新将其拉起，无需运维人员介入。
Kubernetes 可以根据当前 Nebula 集群的资源利用率情况水平伸缩 Nebula 集群，从而提供集群的性能。

下面来讲解下具体的实践内容。

集群部署

硬件和软件要求

这里主要罗列下本文部署涉及到的机器、操作系统参数

操作系统使用的 CentOS-7.6.1810 x86_64
虚拟机配置
- 4 CPU
- 8G 内存
- 50G 系统盘
- 50G 数据盘A
- 50G 数据盘B
Kubernetes 集群版本 v1.16
Nebula 版本为 v1.0.0-rc3
使用本地 PV 作为数据存储

kubernetes 集群规划

以下为集群清单

服务器 IP	nebula 实例	role
192.168.0.1		k8s-master
192.168.0.2	graphd, metad-0, storaged-0	k8s-slave
192.168.0.3	graphd, metad-1, storaged-1	k8s-slave
192.168.0.4	graphd, metad-2, storaged-2	k8s-slave

Kubernetes 待部署组件

安装 Helm
准备本地磁盘，并安装本地卷插件
安装 nebula 集群
安装 ingress-controller

安装 Helm

Helm 是 Kubernetes 集群上的包管理工具，类似 CentOS 上的 yum，Ubuntu 上的 apt-get。使用 Helm 可以极大地降低使用 Kubernetes 部署应用的门槛。由于本篇文章不做 Helm 详细介绍，有兴趣的小伙伴可自行阅读《Helm 入门指南》

下载安装Helm

使用下面命令在终端执行即可安装 Helm

[root@nebula ~]# wget https://get.helm.sh/helm-v3.0.1-linux-amd64.tar.gz  [root@nebula ~]# tar -zxvf helm/helm-v3.0.1-linux-amd64.tgz  [root@nebula ~]# mv linux-amd64/helm /usr/bin/helm  [root@nebula ~]# chmod +x /usr/bin/helm

查看 Helm 版本

执行 helm version 命令即可查看对应的 Helm 版本，以文本为例，以下为输出结果：

version.BuildInfo{      Version:"v3.0.1",      GitCommit:"7c22ef9ce89e0ebeb7125ba2ebf7d421f3e82ffa",      GitTreeState:"clean",      GoVersion:"go1.13.4"  }

设置本地磁盘

在每台机器上做如下配置

创建 mount 目录

[root@nebula ~]# sudo mkdir -p /mnt/disks

格式化数据盘

[root@nebula ~]# sudo mkfs.ext4 /dev/diskA  [root@nebula ~]# sudo mkfs.ext4 /dev/diskB

挂载数据盘

[root@nebula ~]# DISKA_UUID=$(blkid -s UUID -o value /dev/diskA)  [root@nebula ~]# DISKB_UUID=$(blkid -s UUID -o value /dev/diskB)  [root@nebula ~]# sudo mkdir /mnt/disks/$DISKA_UUID  [root@nebula ~]# sudo mkdir /mnt/disks/$DISKB_UUID  [root@nebula ~]# sudo mount -t ext4 /dev/diskA /mnt/disks/$DISKA_UUID  [root@nebula ~]# sudo mount -t ext4 /dev/diskB /mnt/disks/$DISKB_UUID    [root@nebula ~]# echo UUID=`sudo blkid -s UUID -o value /dev/diskA` /mnt/disks/$DISKA_UUID ext4 defaults 0 2 | sudo tee -a /etc/fstab  [root@nebula ~]# echo UUID=`sudo blkid -s UUID -o value /dev/diskB` /mnt/disks/$DISKB_UUID ext4 defaults 0 2 | sudo tee -a /etc/fstab

部署本地卷插件

[root@nebula ~]# curl https://github.com/kubernetes-sigs/sig-storage-local-static-provisioner/archive/v2.3.3.zip  [root@nebula ~]# unzip v2.3.3.zip

修改 v2.3.3/helm/provisioner/values.yaml

#  # Common options.  #  common:    #    # Defines whether to generate service account and role bindings.    #    rbac: true    #    # Defines the namespace where provisioner runs    #    namespace: default    #    # Defines whether to create provisioner namespace    #    createNamespace: false    #    # Beta PV.NodeAffinity field is used by default. If running against pre-1.10    # k8s version, the `useAlphaAPI` flag must be enabled in the configMap.    #    useAlphaAPI: false    #    # Indicates if PVs should be dependents of the owner Node.    #    setPVOwnerRef: false    #    # Provisioner clean volumes in process by default. If set to true, provisioner    # will use Jobs to clean.    #    useJobForCleaning: false    #    # Provisioner name contains Node.UID by default. If set to true, the provisioner    # name will only use Node.Name.    #    useNodeNameOnly: false    #    # Resync period in reflectors will be random between minResyncPeriod and    # 2*minResyncPeriod. Default: 5m0s.    #    #minResyncPeriod: 5m0s    #    # Defines the name of configmap used by Provisioner    #    configMapName: "local-provisioner-config"    #    # Enables or disables Pod Security Policy creation and binding    #    podSecurityPolicy: false  #  # Configure storage classes.  #  classes:  - name: fast-disks # Defines name of storage classe.    # Path on the host where local volumes of this storage class are mounted    # under.    hostDir: /mnt/fast-disks    # Optionally specify mount path of local volumes. By default, we use same    # path as hostDir in container.    # mountDir: /mnt/fast-disks    # The volume mode of created PersistentVolume object. Default to Filesystem    # if not specified.    volumeMode: Filesystem    # Filesystem type to mount.    # It applies only when the source path is a block device,    # and desire volume mode is Filesystem.    # Must be a filesystem type supported by the host operating system.    fsType: ext4    blockCleanerCommand:    #  Do a quick reset of the block device during its cleanup.    #  - "/scripts/quick_reset.sh"    #  or use dd to zero out block dev in two iterations by uncommenting these lines    #  - "/scripts/dd_zero.sh"    #  - "2"    # or run shred utility for 2 iteration.s       - "/scripts/shred.sh"       - "2"    # or blkdiscard utility by uncommenting the line below.    #  - "/scripts/blkdiscard.sh"    # Uncomment to create storage class object with default configuration.    # storageClass: true    # Uncomment to create storage class object and configure it.    # storageClass:      # reclaimPolicy: Delete # Available reclaim policies: Delete/Retain, defaults: Delete.      # isDefaultClass: true # set as default class    #  # Configure DaemonSet for provisioner.  #  daemonset:    #    # Defines the name of a Provisioner    #    name: "local-volume-provisioner"    #    # Defines Provisioner's image name including container registry.    #    image: quay.io/external_storage/local-volume-provisioner:v2.3.3    #    # Defines Image download policy, see kubernetes documentation for available values.    #    #imagePullPolicy: Always    #    # Defines a name of the service account which Provisioner will use to communicate with API server.    #    serviceAccount: local-storage-admin    #    # Defines a name of the Pod Priority Class to use with the Provisioner DaemonSet    #    # Note that if you want to make it critical, specify "system-cluster-critical"    # or "system-node-critical" and deploy in kube-system namespace.    # Ref: https://k8s.io/docs/tasks/administer-cluster/guaranteed-scheduling-critical-addon-pods/#marking-pod-as-critical    #    #priorityClassName: system-node-critical    # If configured, nodeSelector will add a nodeSelector field to the DaemonSet PodSpec.    #    # NodeSelector constraint for local-volume-provisioner scheduling to nodes.    # Ref: https://kubernetes.io/docs/concepts/configuration/assign-pod-node/#nodeselector    nodeSelector: {}    #    # If configured KubeConfigEnv will (optionally) specify the location of kubeconfig file on the node.    #  kubeConfigEnv: KUBECONFIG    #    # List of node labels to be copied to the PVs created by the provisioner in a format:    #    #  nodeLabels:    #    - failure-domain.beta.kubernetes.io/zone    #    - failure-domain.beta.kubernetes.io/region    #    # If configured, tolerations will add a toleration field to the DaemonSet PodSpec.    #    # Node tolerations for local-volume-provisioner scheduling to nodes with taints.    # Ref: https://kubernetes.io/docs/concepts/configuration/taint-and-toleration/    tolerations: []    #    # If configured, resources will set the requests/limits field to the Daemonset PodSpec.    # Ref: https://kubernetes.io/docs/concepts/configuration/manage-compute-resources-container/    resources: {}  #  # Configure Prometheus monitoring  #  prometheus:    operator:      ## Are you using Prometheus Operator?      enabled: false        serviceMonitor:        ## Interval at which Prometheus scrapes the provisioner        interval: 10s          # Namespace Prometheus is installed in        namespace: monitoring          ## Defaults to whats used if you follow CoreOS [Prometheus Install Instructions](https://github.com/coreos/prometheus-operator/tree/master/helm#tldr)        ## [Prometheus Selector Label](https://github.com/coreos/prometheus-operator/blob/master/helm/prometheus/templates/prometheus.yaml#L65)        ## [Kube Prometheus Selector Label](https://github.com/coreos/prometheus-operator/blob/master/helm/kube-prometheus/values.yaml#L298)        selector:          prometheus: kube-prometheus

将hostDir: /mnt/fast-disks 改成hostDir: /mnt/disks

将# storageClass: true 改成 storageClass: true

然后执行：

#安装  [root@nebula ~]# helm install local-static-provisioner v2.3.3/helm/provisioner  #查看local-static-provisioner部署情况  [root@nebula ~]# helm list

部署 nebula 集群

下载 nebula helm-chart 包

# 下载nebula  [root@nebula ~]# wget https://github.com/vesoft-inc/nebula/archive/master.zip  # 解压  [root@nebula ~]# unzip master.zip

设置 Kubernetes slave 节点

下面是 Kubernetes 节点列表，我们需要设置 slave 节点的调度标签。可以将 192.168.0.2，192.168.0.3，192.168.0.4 打上 nebula: "yes" 的标签。

服务器 IP	kubernetes roles	nodeName
192.168.0.1	master	192.168.0.1
192.168.0.2	worker	192.168.0.2
192.168.0.3	worker	192.168.0.3
192.168.0.4	worker	192.168.0.4

具体操作如下：

[root@nebula ~]# kubectl  label node 192.168.0.2 nebula="yes" --overwrite  [root@nebula ~]# kubectl  label node 192.168.0.3 nebula="yes" --overwrite  [root@nebula ~]# kubectl  label node 192.168.0.4 nebula="yes" --overwrite

调整 nebula helm chart 默认的 values 值

nebula helm-chart 包目录如下:

master/kubernetes/  └── helm      ├── Chart.yaml      ├── templates      │   ├── configmap.yaml      │   ├── deployment.yaml      │   ├── _helpers.tpl      │   ├── ingress-configmap.yaml      │   ├── NOTES.txt      │   ├── pdb.yaml      │   ├── service.yaml      │   └── statefulset.yaml      └── values.yaml    2 directories, 10 files

我们需要调整 master/kubernetes/values.yaml 里面的 MetadHosts 的值，将这个 IP List 替换本环境的 3 个 k8s worker 的 ip。

MetadHosts:    - 192.168.0.2:44500    - 192.168.0.3:44500    - 192.168.0.4:44500

通过 helm 安装 nebula

# 安装  [root@nebula ~]# helm install nebula master/kubernetes/helm  # 查看  [root@nebula ~]# helm status nebula  # 查看k8s集群上nebula部署情况  [root@nebula ~]# kubectl get pod  | grep nebula  nebula-graphd-579d89c958-g2j2c                   1/1     Running            0          1m  nebula-graphd-579d89c958-p7829                   1/1     Running            0          1m  nebula-graphd-579d89c958-q74zx                   1/1     Running            0          1m  nebula-metad-0                                   1/1     Running            0          1m  nebula-metad-1                                   1/1     Running            0          1m  nebula-metad-2                                   1/1     Running            0          1m  nebula-storaged-0                                1/1     Running            0          1m  nebula-storaged-1                                1/1     Running            0          1m  nebula-storaged-2                                1/1     Running            0          1m

部署 Ingress-controller

Ingress-controller 是 Kubernetes 的一个 Add-Ons。Kubernetes 通过 ingress-controller 将 Kubernetes 内部署的服务暴露给外部用户访问。Ingress-controller 还提供负载均衡的功能，可以将外部访问流量平摊给 k8s 中应用的不同的副本。

选择一个节点部署 Ingress-controller

[root@nebula ~]# kubectl get node  NAME              STATUS     ROLES    AGE   VERSION  192.168.0.1       Ready      master   82d   v1.16.1  192.168.0.2       Ready      <none>   82d   v1.16.1  192.168.0.3       Ready      <none>   82d   v1.16.1  192.168.0.4       Ready      <none>   82d   v1.16.1  [root@nebula ~]# kubectl label node 192.168.0.4 ingress=yes

编写 ingress-nginx.yaml 部署文件

apiVersion: v1  kind: Namespace  metadata:    name: ingress-nginx    labels:      app.kubernetes.io/name: ingress-nginx      app.kubernetes.io/part-of: ingress-nginx  ---  kind: ConfigMap  apiVersion: v1  metadata:    name: nginx-configuration    namespace: ingress-nginx    labels:      app.kubernetes.io/name: ingress-nginx      app.kubernetes.io/part-of: ingress-nginx  ---  kind: ConfigMap  apiVersion: v1  metadata:    name: tcp-services    namespace: ingress-nginx    labels:      app.kubernetes.io/name: ingress-nginx      app.kubernetes.io/part-of: ingress-nginx  ---  kind: ConfigMap  apiVersion: v1  metadata:    name: udp-services    namespace: ingress-nginx    labels:      app.kubernetes.io/name: ingress-nginx      app.kubernetes.io/part-of: ingress-nginx  ---  apiVersion: v1  kind: ServiceAccount  metadata:    name: nginx-ingress-serviceaccount    namespace: ingress-nginx    labels:      app.kubernetes.io/name: ingress-nginx      app.kubernetes.io/part-of: ingress-nginx    ---  apiVersion: rbac.authorization.k8s.io/v1beta1  kind: ClusterRole  metadata:    name: nginx-ingress-clusterrole    labels:      app.kubernetes.io/name: ingress-nginx      app.kubernetes.io/part-of: ingress-nginx  rules:    - apiGroups:        - ""      resources:        - configmaps        - endpoints        - nodes        - pods        - secrets      verbs:        - list        - watch    - apiGroups:        - ""      resources:        - nodes      verbs:        - get    - apiGroups:        - ""      resources:        - services      verbs:        - get        - list        - watch    - apiGroups:        - "extensions"        - "networking.k8s.io"      resources:        - ingresses      verbs:        - get        - list        - watch    - apiGroups:        - ""      resources:        - events      verbs:        - create        - patch    - apiGroups:        - "extensions"        - "networking.k8s.io"      resources:        - ingresses/status      verbs:        - update  ---  apiVersion: rbac.authorization.k8s.io/v1beta1  kind: Role  metadata:    name: nginx-ingress-role    namespace: ingress-nginx    labels:      app.kubernetes.io/name: ingress-nginx      app.kubernetes.io/part-of: ingress-nginx  rules:    - apiGroups:        - ""      resources:        - configmaps        - pods        - secrets        - namespaces      verbs:        - get    - apiGroups:        - ""      resources:        - configmaps      resourceNames:        # Defaults to "<election-id>-<ingress-class>"        # Here: "<ingress-controller-leader>-<nginx>"        # This has to be adapted if you change either parameter        # when launching the nginx-ingress-controller.        - "ingress-controller-leader-nginx"      verbs:        - get        - update    - apiGroups:        - ""      resources:        - configmaps      verbs:        - create    - apiGroups:        - ""      resources:        - endpoints      verbs:        - get  ---  apiVersion: rbac.authorization.k8s.io/v1beta1  kind: RoleBinding  metadata:    name: nginx-ingress-role-nisa-binding    namespace: ingress-nginx    labels:      app.kubernetes.io/name: ingress-nginx      app.kubernetes.io/part-of: ingress-nginx  roleRef:    apiGroup: rbac.authorization.k8s.io    kind: Role    name: nginx-ingress-role  subjects:    - kind: ServiceAccount      name: nginx-ingress-serviceaccount      namespace: ingress-nginx    ---  apiVersion: rbac.authorization.k8s.io/v1beta1  kind: ClusterRoleBinding  metadata:    name: nginx-ingress-clusterrole-nisa-binding    labels:      app.kubernetes.io/name: ingress-nginx      app.kubernetes.io/part-of: ingress-nginx  roleRef:    apiGroup: rbac.authorization.k8s.io    kind: ClusterRole    name: nginx-ingress-clusterrole  subjects:    - kind: ServiceAccount      name: nginx-ingress-serviceaccount      namespace: ingress-nginx    ---  apiVersion: apps/v1  kind: DaemonSet  metadata:    name: nginx-ingress-controller    namespace: ingress-nginx    labels:      app.kubernetes.io/name: ingress-nginx      app.kubernetes.io/part-of: ingress-nginx  spec:    selector:      matchLabels:        app.kubernetes.io/name: ingress-nginx        app.kubernetes.io/part-of: ingress-nginx    template:      metadata:        labels:          app.kubernetes.io/name: ingress-nginx          app.kubernetes.io/part-of: ingress-nginx        annotations:          prometheus.io/port: "10254"          prometheus.io/scrape: "true"      spec:        hostNetwork: true        tolerations:          - key: "node-role.kubernetes.io/master"            operator: "Exists"            effect: "NoSchedule"        affinity:          podAntiAffinity:            requiredDuringSchedulingIgnoredDuringExecution:              - labelSelector:                  matchExpressions:                    - key: app.kubernetes.io/name                      operator: In                      values:                        - ingress-nginx                topologyKey: "ingress-nginx.kubernetes.io/master"        nodeSelector:          ingress: "yes"        serviceAccountName: nginx-ingress-serviceaccount        containers:          - name: nginx-ingress-controller            image: quay.io/kubernetes-ingress-controller/nginx-ingress-controller-amd64:0.26.1            args:              - /nginx-ingress-controller              - --configmap=$(POD_NAMESPACE)/nginx-configuration              - --tcp-services-configmap=default/graphd-services              - --udp-services-configmap=$(POD_NAMESPACE)/udp-services              - --publish-service=$(POD_NAMESPACE)/ingress-nginx              - --annotations-prefix=nginx.ingress.kubernetes.io              - --http-port=8000            securityContext:              allowPrivilegeEscalation: true              capabilities:                drop:                  - ALL                add:                  - NET_BIND_SERVICE              # www-data -> 33              runAsUser: 33            env:              - name: POD_NAME                valueFrom:                  fieldRef:                    fieldPath: metadata.name              - name: POD_NAMESPACE                valueFrom:                  fieldRef:                    fieldPath: metadata.namespace            ports:              - name: http                containerPort: 80              - name: https                containerPort: 443            livenessProbe:              failureThreshold: 3              httpGet:                path: /healthz                port: 10254                scheme: HTTP              initialDelaySeconds: 10              periodSeconds: 10              successThreshold: 1              timeoutSeconds: 10            readinessProbe:              failureThreshold: 3              httpGet:                path: /healthz                port: 10254                scheme: HTTP              periodSeconds: 10              successThreshold: 1              timeoutSeconds: 10

部署 ingress-nginx

# 部署  [root@nebula ~]# kubectl create -f ingress-nginx.yaml  # 查看部署情况  [root@nebula ~]# kubectl get pod -n ingress-nginx  NAME                             READY   STATUS    RESTARTS   AGE  nginx-ingress-controller-mmms7   1/1     Running   2          1m

访问 nebula 集群

查看 ingress-nginx 所在的节点：

[root@nebula ~]# kubectl get node -l ingress=yes -owide  NAME            STATUS   ROLES    AGE   VERSION   INTERNAL-IP    EXTERNAL-IP   OS-IMAGE                KERNEL-VERSION          CONTAINER-RUNTIME  192.168.0.4     Ready    <none>   1d   v1.16.1    192.168.0.4    <none>        CentOS Linux 7 (Core)   7.6.1810.el7.x86_64     docker://19.3.3

访问 nebula 集群:

[root@nebula ~]# docker run --rm -ti --net=host vesoft/nebula-console:nightly --addr=192.168.0.4 --port=3699

FAQ

如何搭建一套 Kubernetes 集群？

搭建高可用的 Kubernetes 可以参考社区文档：https://kubernetes.io/docs/setup/production-environment/tools/kubeadm/high-availability/

你也可以通过 minikube 搭建本地的 Kubernetes 集群，参考文档：https://kubernetes.io/docs/setup/learning-environment/minikube/

如何调整 nebula 集群的部署参数?

在使用 helm install 时，使用 –set 可以设置部署参数，从而覆盖掉 helm chart 中 values.yaml 中的变量。参考文档：https://helm.sh/docs/intro/using_helm/

如何查看 nebula 集群状况？

使用kubectl get pod | grep nebula命令，或者直接在 Kubernetes dashboard 上查看 nebula 集群的运行状况。

如何使用其他类型的存储？

参考文档：https://kubernetes.io/zh/docs/concepts/storage/storage-classes/

参考资料

附录

Nebula Graph：一个开源的分布式图数据库
GitHub：https://github.com/vesoft-inc/nebula
知乎：zhihu.com/org/nebulagraph/posts
微博：weibo.com/nebulagraph