Prometheus监控K8S–通过kube-prometheus
1、安装之前，查看对应的版本信息

由于安装的是1.21版本所以需要下载release-0.9
2、下载对应的版本
 https://github.com/prometheus-operator/kube-prometheus

CSS
git clone  https://github.com/prometheus-operator/kube-prometheus.git
[root@k8s-master prometheus]# cd kube-prometheus
[root@k8s-master kube-prometheus]# ls
build.sh code-of-conduct.md developer-workspace example.jsonnet experimental go.sum jsonnetfile.json kubescape-exceptions.json LICENSE manifests README.md scripts
CHANGELOG.md CONTRIBUTING.md docs examples go.mod jsonnet jsonnetfile.lock.json kustomization.yaml Makefile nohup.out RELEASE.md tests

SQL
1、安装之前先导入镜像，
[root@k8s-master images]# ll
total 502752
-rw-r–r-- 1 root root 61165568 Apr 21 20:43 alertmanager.tar
-rw-r–r-- 1 root root 285516288 Apr 21 20:43 grafana.tar
-rw-r–r-- 1 root root 62647808 Apr 21 20:43 kube-rbac-proxy.tar
-rwxr-xr-x 1 root root 205 Apr 21 20:46 load-images.sh
-rw------- 1 root root 22312960 Apr 21 16:58 node-exporter.tar
-rw-r–r-- 1 root root 69536256 Apr 21 20:43 prometheus-adapter.tar
-rw-r–r-- 1 root root 13626368 Apr 21 20:43 prometheus-config-reloader.tar
[root@k8s-master images]# sh load-images.sh

2、安装命令
kubectl create -f manifests/setup

3、待定前面的容器启动后执行
kubectl create -f manifests/

4、查看资源
[root@k8s-master ~]# kubectl get pod -nmonitoring -owide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
alertmanager-main-0 2/2 Running 0 159m 10.244.2.85 k8s-node2 <none> <none>
alertmanager-main-1 2/2 Running 0 159m 10.244.3.101 k8s-node1 <none> <none>
alertmanager-main-2 2/2 Running 0 159m 10.244.4.69 k8s-node6 <none> <none>
blackbox-exporter-5cb5d7479d-5bvfh 3/3 Running 0 160m 10.244.2.83 k8s-node2 <none> <none>
grafana-7ccd96765c-dzpqs 1/1 Running 0 160m 10.244.2.84 k8s-node2 <none> <none>
kube-state-metrics-79f478884f-mckgc 3/3 Running 0 160m 10.244.4.67 k8s-node6 <none> <none>
node-exporter-54zt8 2/2 Running 14 160m 172.16.4.169 node <none> <none>
node-exporter-586pr 2/2 Running 10 160m 172.16.4.203 k8s-node2 <none> <none>
node-exporter-bqcxx 2/2 Running 0 160m 172.16.4.121 k8s-node1 <none> <none>
node-exporter-j2h6j 2/2 Running 14 160m 172.16.7.210 k8s-node6 <none> <none>
prometheus-adapter-7bf7ff5b67-m9hnm 1/1 Running 0 160m 10.244.3.99 k8s-node1 <none> <none>
prometheus-adapter-7bf7ff5b67-t8tct 1/1 Running 0 160m 10.244.3.100 k8s-node1 <none> <none>
prometheus-k8s-0 2/2 Running 0 159m 10.244.2.86 k8s-node2 <none> <none>
prometheus-k8s-1 2/2 Running 0 159m 10.244.4.70 k8s-node6 <none> <none>
prometheus-operator-b998f8597-zq797 2/2 Running 0 160m 10.244.4.68 k8s-node6 <none> <none>
3、卸载

PowerShell
kubectl delete --ignore-not-found=true -f manifests/ -f manifests/setup

4、通过web访问grafana

Apache
nohub kubectl port-forward --address 0.0.0.0 svc/grafana 3000:3000 -n monitoring
5、在grafana里导入相关的dashboard
依次点击+ --> Import–导入相关的json文件

具体页面如下：