使用 Sealos + Longhorn 部署 KubeSphere v3.0.0

原创
09/29 23:50
阅读数 423

本文来自 KubeSphere 社区用户 Will,演示如何使用 Sealos + Longhorn 部署一个带有持久化存储的 Kubernetes 集群,然后使用 ks-installer 在该集群上部署 KubeSphere 3.0.0。这是一篇最适合小白初次上手的 KubeSphere 3.0.0 快速部署和体验的文章🚀。

Sealos 简介

Sealos (https://sealyun.com/),只能用丝滑一词形容的 Kubernetes 高可用安装工具,一条命令,离线安装,包含所有依赖,内核负载不依赖 haproxy keepalived,纯 Golang 开发,99 年证书,支持 v1.16 ~ v1.19。

Longhorn简介

Longhornhttps://www.rancher.cn/longhorn)是 Rancher 开源的 Kubernetes 高可用持久化存储,提供简单的增量快照和备份,支持跨集群灾难恢复。

KubeSphere 简介

KubeSphere(https://kubesphere.io是在 Kubernetes 之上构建的以应用为中心的多租户容器平台,完全开源,支持多云与多集群管理,提供全栈的 IT 自动化运维的能力,简化企业的 DevOps 工作流。KubeSphere 提供了运维友好的向导式操作界面,帮助企业快速构建一个强大和功能丰富的容器云平台。

KubeSphere 支持如下两种安装方式:

  • 使用 KubeKey 部署 Kubernetes 集群 + KubeSphere
  • 在已有 Kubernetes 集群部署 KubeSphere

对于已有 Kubernetes 集群的用户来说,在已有 Kubernetes 集群部署 KubeSphere 具有更高的灵活性。下面演示单独部署一个 Kubernetes 集群,并在集群上部署 KubeSphere。

使用 Sealos 部署 Kubernetes 集群

准备 4 个节点,由于实验的机器有限,我们暂时准备 3 个 master 和 1 个 node,注意在实际的生产环境建议配置 3 master 和至少 3 node。所有节点必须配置主机名,并确认节点时间同步:

hostnamectl set-hostname xx
yum install -y chrony
systemctl enable --now chronyd
timedatectl set-timezone Asia/Shanghai

在第一个 master 节点操作,下载部署工具及离线包:

# 基于 go 的二进制安装程序
wget -c https://sealyun.oss-cn-beijing.aliyuncs.com/latest/sealos && \
    chmod +x sealos && mv sealos /usr/bin
# 以 K8s v1.18.8为例,不建议使用 v1.19.x,因为 KubeSphere v3.0.0 暂不支持
wget -c https://sealyun.oss-cn-beijing.aliyuncs.com/cd3d5791b292325d38bbfaffd9855312-1.18.8/kube1.18.8.tar.gz

执行以下命令部署 Kubernetes 集群,passwd 为所有节点 root 密码:

sealos init --passwd 123456 \
  --master 10.39.140.248 \
  --master 10.39.140.249 \
  --master 10.39.140.250 \
  --node 10.39.140.251 \
  --pkg-url kube1.18.8.tar.gz \
  --version v1.18.8

确认 Kubernetes 集群运行正常:

# kubectl get nodes
NAME          STATUS   ROLES    AGE   VERSION
k8s-master1   Ready    master   13h   v1.18.8
k8s-master2   Ready    master   13h   v1.18.8
k8s-master3   Ready    master   13h   v1.18.8
k8s-node1     Ready    <none>   13h   v1.18.8

部署 Longhorn 存储

Longhorn 推荐单独挂盘作为存储使用,这里作为测试直接使用本地存储目录 /data/longhorn,默认为 /var/lib/longhorn

注意,KubeSphere 有几个组件申请的 PV 大小为 20G,确保节点空间充足,否则可能出现 PV 能够绑定成功但没有满足条件的节点可调度的情况。

安装具有 3 数据副本的 Longhorn 至少需要 3 个节点,这里去除 master 节点污点使其可调度 Pod:

kubectl taint nodes --all node-role.kubernetes.io/master-  

在 k8s-master1 安装 Helm:

version=v3.3.1
curl -LO https://repo.huaweicloud.com/helm/${version}/helm-${version}-linux-amd64.tar.gz
tar -zxvf helm-${version}-linux-amd64.tar.gz
mv linux-amd64/helm /usr/local/bin/helm && rm -rf linux-amd64

所有节点安装 longhorn 依赖:

yum install -y iscsi-initiator-utils
systemctl enable --now iscsid

添加 Longhorn Chart,如果网络较差可以去 Longhorn 的 github release 下载 Chart:

helm repo add longhorn https://charts.longhorn.io
helm repo update

部署 Longhorn,支持离线部署,需要提前推送镜像到私有仓库 longhorn.io 下:

kubectl create namespace longhorn-system

helm install longhorn \
  --namespace longhorn-system \
  --set defaultSettings.defaultDataPath="/data/longhorn/" \
  --set defaultSettings.defaultReplicaCount=3 \
  --set service.ui.type=NodePort \
  --set service.ui.nodePort=30890 \
  #--set privateRegistry.registryUrl=10.39.140.196:8081 \
  longhorn/longhorn

确认 Longhorn 运行正常:

[root@jenkins longhorn]# kubectl -n longhorn-system get pods
NAME                                        READY   STATUS    RESTARTS   AGE
csi-attacher-58b856dcff-9kqdt               1/1     Running   0          13h
csi-attacher-58b856dcff-c4zzp               1/1     Running   0          13h
csi-attacher-58b856dcff-tvfw2               1/1     Running   0          13h
csi-provisioner-56dd9dc55b-6ps8m            1/1     Running   0          13h
csi-provisioner-56dd9dc55b-m7gz4            1/1     Running   0          13h
csi-provisioner-56dd9dc55b-s9bh4            1/1     Running   0          13h
csi-resizer-6b87c4d9f8-2skth                1/1     Running   0          13h
csi-resizer-6b87c4d9f8-sqn2g                1/1     Running   0          13h
csi-resizer-6b87c4d9f8-z6xql                1/1     Running   0          13h
engine-image-ei-b99baaed-5fd7m              1/1     Running   0          13h
engine-image-ei-b99baaed-jcjxj              1/1     Running   0          12h
engine-image-ei-b99baaed-n6wxc              1/1     Running   0          12h
engine-image-ei-b99baaed-qxfhg              1/1     Running   0          12h
instance-manager-e-44ba7ac9                 1/1     Running   0          12h
instance-manager-e-48676e4a                 1/1     Running   0          12h
instance-manager-e-57bd994b                 1/1     Running   0          12h
instance-manager-e-753c704f                 1/1     Running   0          13h
instance-manager-r-4f4be1c1                 1/1     Running   0          12h
instance-manager-r-68bfb49b                 1/1     Running   0          12h
instance-manager-r-ccb87377                 1/1     Running   0          12h
instance-manager-r-e56429be                 1/1     Running   0          13h
longhorn-csi-plugin-fqgf7                   2/2     Running   0          12h
longhorn-csi-plugin-gbrnf                   2/2     Running   0          13h
longhorn-csi-plugin-kjj6b                   2/2     Running   0          12h
longhorn-csi-plugin-tvbvj                   2/2     Running   0          12h
longhorn-driver-deployer-74bb5c9fcb-khmbk   1/1     Running   0          14h
longhorn-manager-82ztz                      1/1     Running   0          12h
longhorn-manager-8kmsn                      1/1     Running   0          12h
longhorn-manager-flmfl                      1/1     Running   0          12h
longhorn-manager-mz6zj                      1/1     Running   0          14h
longhorn-ui-77c6d6f5b7-nzsg2                1/1     Running   0          14h

确认默认的 StorageClass 已就绪:

# kubectl get sc
NAME                 PROVISIONER          RECLAIMPOLICY   VOLUMEBINDINGMODE   ALLOWVOLUMEEXPANSION   AGE
longhorn (default)   driver.longhorn.io   Delete          Immediate           true                   14h

登录 Longhorn UI 确认节点处于可调度状态:

Longhorn UI

Longhorn UI 查看绑定的 PV 卷

longhorn UI 查看绑定的 PV 卷

查看存储卷详情

查看存储卷详情

在 Kubernetes 上部署 KubeSphere

使用 ks-installer 项目来安装 KubeSphere,下载 KubeSphere 安装的 Yaml 文件:

wget https://raw.githubusercontent.com/kubesphere/ks-installer/v3.0.0/deploy/kubesphere-installer.yaml
wget https://raw.githubusercontent.com/kubesphere/ks-installer/v3.0.0/deploy/cluster-configuration.yaml

KubeSphere 默认仅开启了最小化安装,可修改 cluster-configuration.yaml,找到相应字段开启需要安装的功能组件,以下仅为参考:

  devops:
    enabled: true
    ......
  logging:
    enabled: true
    ......
  metrics_server:
    enabled: true
    ......
  openpitrix:
    enabled: true
    ......

执行命令部署 KubeSphere:

kubectl apply -f kubesphere-installer.yaml
kubectl apply -f cluster-configuration.yaml

查看部署日志,确认无报错:

kubectl logs -n kubesphere-system $(kubectl get pod -n kubesphere-system -l app=ks-install -o jsonpath='{.items[0].metadata.name}') -f

部署完成后确认所有 KubeSphere 相关的 Pod 运行正常:

[root@k8s-master1 ~]# kubectl get pods -A | grep kubesphere
kubesphere-controls-system     default-http-backend-857d7b6856-q24v2                             1/1     Running     0          12h
kubesphere-controls-system     kubectl-admin-58f985d8f6-jl9bj                                    1/1     Running     0          11h
kubesphere-controls-system     kubesphere-router-demo-ns-6c97d4968b-njgrc                        1/1     Running     1          154m
kubesphere-devops-system       ks-jenkins-54455f5db8-hm6kc                                       1/1     Running     0          11h
kubesphere-devops-system       s2ioperator-0                                                     1/1     Running     1          11h
kubesphere-devops-system       uc-jenkins-update-center-cd9464fff-qnvfz                          1/1     Running     0          12h
kubesphere-logging-system      elasticsearch-logging-curator-elasticsearch-curator-160079hmdmb   0/1     Completed   0          11h
kubesphere-logging-system      elasticsearch-logging-data-0                                      1/1     Running     0          12h
kubesphere-logging-system      elasticsearch-logging-data-1                                      1/1     Running     0          12h
kubesphere-logging-system      elasticsearch-logging-discovery-0                                 1/1     Running     0          12h
kubesphere-logging-system      fluent-bit-c45h2                                                  1/1     Running     0          12h
kubesphere-logging-system      fluent-bit-kptfc                                                  1/1     Running     0          12h
kubesphere-logging-system      fluent-bit-rzjfp                                                  1/1     Running     0          12h
kubesphere-logging-system      fluent-bit-wztkp                                                  1/1     Running     0          12h
kubesphere-logging-system      fluentbit-operator-855d4b977d-fk6hs                               1/1     Running     0          12h
kubesphere-logging-system      ks-events-exporter-5bc4d9f496-x297f                               2/2     Running     0          12h
kubesphere-logging-system      ks-events-operator-8dbf7fccc-9qmml                                1/1     Running     0          12h
kubesphere-logging-system      ks-events-ruler-698b7899c7-fkn4l                                  2/2     Running     0          12h
kubesphere-logging-system      ks-events-ruler-698b7899c7-hw6rq                                  2/2     Running     0          12h
kubesphere-logging-system      logsidecar-injector-deploy-74c66bfd85-cxkxm                       2/2     Running     0          12h
kubesphere-logging-system      logsidecar-injector-deploy-74c66bfd85-lzxbm                       2/2     Running     0          12h
kubesphere-monitoring-system   alertmanager-main-0                                               2/2     Running     0          11h
kubesphere-monitoring-system   alertmanager-main-1                                               2/2     Running     0          11h
kubesphere-monitoring-system   alertmanager-main-2                                               2/2     Running     0          11h
kubesphere-monitoring-system   kube-state-metrics-95c974544-r8kmq                                3/3     Running     0          12h
kubesphere-monitoring-system   node-exporter-9ddxn                                               2/2     Running     0          12h
kubesphere-monitoring-system   node-exporter-dw929                                               2/2     Running     0          12h
kubesphere-monitoring-system   node-exporter-ht868                                               2/2     Running     0          12h
kubesphere-monitoring-system   node-exporter-nxdsm                                               2/2     Running     0          12h
kubesphere-monitoring-system   notification-manager-deployment-7c8df68d94-hv56l                  1/1     Running     0          12h
kubesphere-monitoring-system   notification-manager-deployment-7c8df68d94-ttdsg                  1/1     Running     0          12h
kubesphere-monitoring-system   notification-manager-operator-6958786cd6-pllgc                    2/2     Running     0          12h
kubesphere-monitoring-system   prometheus-k8s-0                                                  3/3     Running     1          11h
kubesphere-monitoring-system   prometheus-k8s-1                                                  3/3     Running     1          11h
kubesphere-monitoring-system   prometheus-operator-84d58bf775-5rqdj                              2/2     Running     0          12h
kubesphere-system              etcd-65796969c7-whbzx                                             1/1     Running     0          12h
kubesphere-system              ks-apiserver-b4dbcc67-2kknm                                       1/1     Running     0          11h
kubesphere-system              ks-apiserver-b4dbcc67-k6jr2                                       1/1     Running     0          11h
kubesphere-system              ks-apiserver-b4dbcc67-q8845                                       1/1     Running     0          11h
kubesphere-system              ks-console-786b9846d4-86hxw                                       1/1     Running     0          12h
kubesphere-system              ks-console-786b9846d4-l6mhj                                       1/1     Running     0          12h
kubesphere-system              ks-console-786b9846d4-wct8z                                       1/1     Running     0          12h
kubesphere-system              ks-controller-manager-7fd8799789-478ks                            1/1     Running     0          11h
kubesphere-system              ks-controller-manager-7fd8799789-hwgmp                            1/1     Running     0          11h
kubesphere-system              ks-controller-manager-7fd8799789-pdbch                            1/1     Running     0          11h
kubesphere-system              ks-installer-64ddc4b77b-c7qz8                                     1/1     Running     0          12h
kubesphere-system              minio-7bfdb5968b-b5v59                                            1/1     Running     0          12h
kubesphere-system              mysql-7f64d9f584-kvxcb                                            1/1     Running     0          12h
kubesphere-system              openldap-0                                                        1/1     Running     0          12h
kubesphere-system              openldap-1                                                        1/1     Running     0          12h
kubesphere-system              redis-ha-haproxy-5c6559d588-2rt6v                                 1/1     Running     9          12h
kubesphere-system              redis-ha-haproxy-5c6559d588-mhj9p                                 1/1     Running     8          12h
kubesphere-system              redis-ha-haproxy-5c6559d588-tgpjv                                 1/1     Running     11         12h
kubesphere-system              redis-ha-server-0                                                 2/2     Running     0          12h
kubesphere-system              redis-ha-server-1                                                 2/2     Running     0          12h
kubesphere-system              redis-ha-server-2                                                 2/2     Running     0          12h

KubeSphere 部分组件使用 Helm 部署,检查 Chart 部署情况:

[root@k8s-master1 ~]# helm ls -A | grep kubesphere
elasticsearch-logging           kubesphere-logging-system       1               2020-09-23 00:49:08.526873742 +0800 CST deployed        elasticsearch-1.22.1            6.7.0-0217                  
elasticsearch-logging-curator   kubesphere-logging-system       1               2020-09-23 00:49:16.117842593 +0800 CST deployed        elasticsearch-curator-1.3.3     5.5.4-0217                  
ks-events                       kubesphere-logging-system       1               2020-09-23 00:51:45.529430505 +0800 CST deployed        kube-events-0.1.0               0.1.0                       
ks-jenkins                      kubesphere-devops-system        1               2020-09-23 01:03:15.106022826 +0800 CST deployed        jenkins-0.19.0                  2.121.3-0217                
ks-minio                        kubesphere-system               2               2020-09-23 00:48:16.990599158 +0800 CST deployed        minio-2.5.16                    RELEASE.2019-08-07T01-59-21Z
ks-openldap                     kubesphere-system               1               2020-09-23 00:03:28.767712181 +0800 CST deployed        openldap-ha-0.1.0               1.0                         
ks-redis                        kubesphere-system               1               2020-09-23 00:03:19.439784188 +0800 CST deployed        redis-ha-3.9.0                  5.0.5                       
logsidecar-injector             kubesphere-logging-system       1               2020-09-23 00:51:57.519733074 +0800 CST deployed        logsidecar-injector-0.1.0       0.1.0                       
notification-manager            kubesphere-monitoring-system    1               2020-09-23 00:54:14.662762759 +0800 CST deployed        notification-manager-0.1.0      0.1.0                       
uc                              kubesphere-devops-system        1               2020-09-23 00:51:37.885154574 +0800 CST deployed        jenkins-update-center-0.8.0     3.0.0    

获取 KubeSphere Console 监听端口,默认为 30880:

kubectl get svc/ks-console -n kubesphere-system

默认登录账号为 admin/P@88w0rd,登录 KubeSphere Console:

在 KubeSphere 查看 Kubernetes 集群总览(界面非常清新简洁):

集群管理总览页

查看 Kubernetes 集群节点信息:

集群节点信息

查看 KubeSphere 服务组件信息:

服务组件信息

访问 KubeSphere 应用商店:

KubeSphere 应用商店

查看 KubeSphere 项目资源:

KubeSphere 项目资源

提示:关于如何在 KubeSphere 平台导入多集群、创建项目与集群资源、开启可插拔功能组件以及创建 CI/CD 流水线,可以参考 KubeSphere 官方文档 (kubesphere.io/docs) 了解更多信息。

清理 KubeSphere 集群

wget https://raw.githubusercontent.com/kubesphere/ks-installer/master/scripts/kubesphere-delete.sh
sh kubesphere-delete.sh

原文首发于:https://blog.csdn.net/networken/article/details/105664147


 KubeSphere

KubeSphere https://kubesphere.io Kubernetes  IT  DevOps 

KubeSphere  Aqara 寿、中国太平保险、RadoreZaloPay KubeSphere Kubernetes DevOps (CI/CD) (Service Mesh)GPU support 


本文分享自微信公众号 - KubeSphere(gh_4660e44db839)。
如有侵权,请联系 support@oschina.cn 删除。
本文参与“OSC源创计划”,欢迎正在阅读的你也加入,一起分享。

展开阅读全文
打赏
0
0 收藏
分享
加载中
更多评论
打赏
0 评论
0 收藏
0
分享
返回顶部
顶部