kubernetes clusterr monitor prometheus+Grafana集群资源监控

tsdb 时序数据库 默认保存60天的数据，若需要持久保存，需要额外的添加外部存储
prometheus主动拉取jobs/exporters数据
prometheus主动拉取pushgateway临时数据 ，short-lived jobs主动推临时数据至pushgateway
alertmanager 报警行为，独立组件，不属于prometheus
systemctl enable —now SERVICESNAME
 
 
 
1、监控指标
  • 集群监控
    • 节点资源利用率
    • 节点数量
    • 运行多少个pod
  • Pod监控
    • 容器指标
    • 应用程序
2、监控平台 prometheus+Grafana
  • prometheus：定时抓取被监控组件状态数据，通过http协议进行系统监控，将收集的数据进行存储基于数据库的方式使用（适用虚拟化，容器化环境）
    • 开源
    • 监控、报警、数据库
    • 以HTTP协议周期性抓取被监控组件状态
    • 不需要复杂的集成过程，使用http接口接入即可
  • Grafana：通过prometheus抓取的数据进行展示，可视化工具
    • 开源的数据分析和可视化工具
    • 支持多种数据源
notion image
3、搭建监控平台
  • 部署prometheus
    • 部署守护进程 node-exporter.yaml
    • rbac-setup.yaml 设置权限
    • configmap.yaml 存储
    • prometheus.deploy.yml 部署过程
    • prometheus.svc.yml 暴露端口
  • 部署grafana
    • grafana-deploy 部署过程
    • grafana-svc 暴露端口
    • grafana-ing 使用ingress

补充

GitHub - prometheus-operator/kube-prometheus at release-0.5
Note that everything is experimental and may change significantly at any time. This repository collects Kubernetes manifests, Grafana dashboards, and Prometheus rules combined with documentation and scripts to provide easy to operate end-to-end Kubernetes cluster monitoring with Prometheus using the Prometheus Operator. The content of this project is written in jsonnet.
GitHub - prometheus-operator/kube-prometheus at release-0.5
https://github.com/prometheus-operator/kube-prometheus/tree/release-0.5
GitHub - prometheus-operator/kube-prometheus at release-0.5
模板
Node Exporter Full dashboard for Grafana | Grafana Labs
Nearly all default values exported by Prometheus node exporter graphed. Only requires the default job_name: node, add as many targets as you need in '/etc/prometheus/prometheus.yml'. - job_name: node static_configs: - targets: ['localhost:9100'] Recommended for prometheus-node-exporter the arguments '--collector.systemd --collector.processes' because the graph uses some of their metrics.
Node Exporter Full dashboard for Grafana | Grafana Labs
https://grafana.com/grafana/dashboards/1860
Node Exporter Full dashboard for Grafana | Grafana Labs
操作指南
安装部署kube-prometheus_周周周周和的博客-CSDN博客_kube-prometheus 部署
prometheus-operator 用于监控kubernetes集群 https://github.com/prometheus-operator/kube-prometheus/archive/main.zip 下载完成后解压 kube-prometheus-main.zip 改完如下图， k8s.gcr.io和 quay.io国内被墙，需要替代方案， 修改prometheus和grafana的持久化方式，默认持久化方式为 emptyDir，最好修改为 pvc，参考 该作者博客 修改 grafana-deployment.yaml的 grafana-storage，创建一个 pvc替换上去 修改 prometheus-prometheus.yaml的 CRD添加 PVC存储 prometheus-service.yaml alertmanager-service.yaml grafana-service.yaml 安装prometheus-operator kubectl apply -f setup/ 查看状态，到 Running 再进行下一步 kubectl get pods -n monitoring 安装prometheus, alertmanager, grafana, kube-state-metrics, node-exporter等资源 grafana 帐号密码 admin/admin kubectl delete --ignore-not-found=true -f manifests/ -f
安装部署kube-prometheus_周周周周和的博客-CSDN博客_kube-prometheus 部署
https://blog.csdn.net/qq850482461/article/details/118759875
安装部署kube-prometheus_周周周周和的博客-CSDN博客_kube-prometheus 部署
notion image
Loading...
目录