k8s集群监控方案--node-exporter+prometheus+grafana
目录
前置条件
一、下载yaml文件
二、部署yaml各个组件
2.1 node-exporter.yaml
2.2 Prometheus
2.3 grafana
2.4访问测试
三、grafana初始化
3.1加载数据源
3.2导入模板
四、helm方式部署
前置条件
安装好k8s集群（几个节点都可以，本人为了方便实验k8s集群只有一个master节点），注意prometheus是部署在k8s集群内部的，不同于传统监控分为监控端和被控端。
| 11.0.1.12 |
k8s-master / node-exporter+prometheus+grafana |
一、下载yaml文件
链接：
解压
[root@prometheus opt]# yum install -y zip unzip tree
[root@prometheus opt]# unzip k8s-prometheus-grafana-master.zip [root@k8s-master k8s-prometheus-grafana-master]# pwd
/opt/k8s-prometheus-grafana-master
[root@k8s-master k8s-prometheus-grafana-master]# tree
.
├── grafana
│?? ├── grafana-deploy.yaml
│?? ├── grafana-ing.yaml
│?? └── grafana-svc.yaml
├── node-exporter.yaml
├── prometheus
│?? ├── configmap.yaml
│?? ├── prometheus.deploy.yml
│?? ├── prometheus.svc.yml
│?? └── rbac-setup.yaml
└── README.md
二、部署yaml各个组件
kubectl命令tab补全
[root@k8s-master ~]# yum install -y bash-completion
[root@k8s-master ~]# source <(kubectl completion bash)
2.1 node-exporter.yaml
[root@k8s-master k8s-prometheus-grafana-master]# kubectl apply -f node-exporter.yaml
daemonset.apps/node-exporter created
service/node-exporter created
#因为只有一个节点，这里如果有多个节点，节点数=node-exporterPod数
[root@k8s-master k8s-prometheus-grafana-master]# kubectl get pod -A | grep node-exporter
kube-system node-exporter-kpdxh 0/1 ContainerCreating 0 28s
[root@k8s-master k8s-prometheus-grafana-master]# kubectl get daemonset -A | grep exporter
kube-system node-exporter 1 1 1 1 1 2m43s
[root@k8s-master k8s-prometheus-grafana-master]# kubectl get service -A | grep exporter
kube-system node-exporter NodePort 10.96.73.86 9100:31672/TCP 2m59s
2.2 Prometheus
[root@k8s-master prometheus]# pwd
/opt/k8s-prometheus-grafana-master/prometheus
[root@k8s-master prometheus]# ls
configmap.yaml prometheus.deploy.yml prometheus.svc.yml rbac-setup.yaml按照顺序 rbac-setup.yaml configmap.yaml prometheus.deploy.yml prometheus.svc.yml ，yaml、yml文件没区别，不用在意。
[root@k8s-master prometheus]# kubectl apply -f rbac-setup.yaml
clusterrole.rbac.authorization.k8s.io/prometheus created
serviceaccount/prometheus created
clusterrolebinding.rbac.authorization.k8s.io/prometheus created
[root@k8s-master prometheus]# kubectl apply -f configmap.yaml
configmap/prometheus-config created
[root@k8s-master prometheus]# kubectl apply -f prometheus.deploy.yml
deployment.apps/prometheus created
[root@k8s-master prometheus]# kubectl apply -f prometheus.svc.yml
service/prometheus created2.3 grafana
[root@k8s-master grafana]# pwd
/opt/k8s-prometheus-grafana-master/grafana
[root@k8s-master grafana]# ls
grafana-deploy.yaml grafana-ing.yaml grafana-svc.yaml按照顺序安装 grafana-deploy.yaml grafana-svc.yaml grafana-ing.yaml
[root@k8s-master grafana]# kubectl apply -f grafana-deploy.yaml
deployment.apps/grafana-core created
[root@k8s-master grafana]# kubectl apply -f grafana-svc.yaml
service/grafana created
[root@k8s-master grafana]# kubectl apply -f grafana-ing.yaml
ingress.extensions/grafana created检查三个pod（node-exporter可能有多个）
[root@k8s-master grafana]# kubectl get pod -A | grep node-exporter
kube-system node-exporter-kpdxh 1/1 Running 0 13m
[root@k8s-master grafana]# kubectl get pod -A | grep prometheus
kube-system prometheus-7486bf7f4b-xb4t8 1/1 Running 0 6m21s
[root@k8s-master grafana]# kubectl get pod -A | grep grafana
kube-system grafana-core-664b68875b-fhjvt 1/1 Running 0 2m18s
检查服务service，三个svc类型均为NodePort
[root@k8s-master grafana]# kubectl get svc -A
NAMESPACE NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
default kubernetes ClusterIP 10.96.0.1 443/TCP 7h17m
kube-system grafana NodePort 10.107.115.11 3000:31748/TCP 3m41s
kube-system kube-dns ClusterIP 10.96.0.10 53/UDP,53/TCP,9153/TCP 7h17m
kube-system node-exporter NodePort 10.96.73.86 9100:31672/TCP 15m
kube-system prometheus NodePort 10.111.178.83 9090:30003/TCP 7m57s
2.4访问测试
curl访问测试
[root@k8s-master grafana]# curl 127.0.0.1:31672
......
#node-exporter收集到的数据
[root@k8s-master grafana]# curl 127.0.0.1:31672/metrics
# HELP go_gc_duration_seconds A summary of the pause duration of garbage collection cycles.
[root@k8s-master grafana]# curl 127.0.0.1:30003
Found.
[root@k8s-master grafana]# curl 127.0.0.1:31748
Found.
访问11.0.1.12:31672/metrics，node-exporter收集到的数据
访问11.0.1.12:30003/graph
选择 "status"--"targets"
三、grafana初始化
访问11.0.1.12:31748/login
账号密码都是admin
3.1加载数据源
监听地址 10.111.178.83:9090，prometheus的service地址
[root@k8s-master grafana]# kubectl get svc -A | grep prometheus
kube-system prometheus NodePort 10.111.178.83 9090:30003/TCP 40m 
3.2导入模板
模板编号315
选择数据源prometheus
完成监控
四、helm方式部署
......（待更新）