Prometheus的前身是Google Borgmon
go语言编写，性能高。
时序数据库
Prometheus支持的监控类型： https://prometheus.io/docs/instrumenting/exporters/
通常用Grafana作为Prometheus的UI
Prometheus的环境搭建
Centos7 的环境
下载最新版本的安装包 wget https://github.com/prometheus/prometheus/releases/download/v2.18.1/prometheus-2.18.1.linux-amd64.tar.gz 解压 tar zxvf prometheus-2.18.1.linux-amd64.tar.gz 启动 ./prometheus –config.file=prometheus.yml
./prometheus –config.file=prometheus.yml –web.enable-lifecycle
动态更新prometheus的配置 curl -X POST http://localhost:9090/-/reload
localhost:9090/graph localhost:9090/metrics
lsof -i tcp:9090
存储原理参考： http://coreos.com/blog/prometheus-2.0-storage-layer-optimization
Grafana
Grafana官网下载地址： https://grafana.com/grafana/download centos7安装： wget https://dl.grafana.com/oss/release/grafana-6.7.3-1.x86_64.rpm sudo yum install grafana-6.7.3-1.x86_64.rpm service grafana-server start
web默认权限 用户名密码都是admin
在web界面中 1、先添加数据源 add data sources -> Prometheus -> 修改Name和URL 2、创建dashboard 保存，起个名字 Add Query Query选择 Prometheus-localhost metrics： prometheus_http_request_duration_seconds_bucket{job=”prometheus”}
Prometheus 报警 AlertManager下载 centos7 wget https://github.com/prometheus/alertmanager/releases/download/v0.20.0/alertmanager-0.20.0.linux-amd64.tar.gz 端口 9093
docker pull prom/prometheus:latest docker pull prom/node-exporter:latest docker pull prom/prometheus:latest
kubectl create -f namespace.yaml kubectl get namespaces kubectl get pods -n ns-monitor
kubectl create -f node-exporter.yaml kubectl get pods -n ns-monitor
访问Node的IP加31672端口 http://192.168.99.101:31672/metrics
kubectl create -f prometheus.yaml kubectl get pod -n ns-monitor
kubectl get svc -n ns-monitor
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
node-exporter-service NodePort 10.100.85.120
验证 http://192.168.99.101:30824/graph
kubectl apply -f grafana.yaml
kubectl get svc -n ns-monitor
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
grafana-service NodePort 10.110.89.59
验证 http://192.168.99.101:31987/graph 默认用户名和密码：admin/admin
kubectl get pods -n ns-monitor -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
grafana-576db894c6-sj4st 1/1 Running 0 12m 172.17.0.13 minikube
http://172.17.0.12:9090
完整安装prometheus+grafana
在需要安装prometheus的node上执行，我这里是192.168.11.12: wget https://github.com/prometheus/prometheus/releases/download/v2.18.1/prometheus-2.18.1.linux-amd64.tar.gz
tar zxvf prometheus-2.18.1.linux-amd64.tar.gz
mkdir -p /etc/prometheus cp prometheus-2.18.1.linux-amd64/* /etc/prometheus/ mv prometheus-2.18.1.linux-amd64/prometheus /bin/
docker pull grafana/grafana:latest
装nfs： yum install nfs-utils rpcbind -y mkdir -p /data/k8s systemctl start rpcbind systemctl enable rpcbind systemctl enable nfs-server.service echo “/data/k8s 192.168.11.0/24(rw,no_root_squash,sync)” »/etc/exports systemctl start nfs-server.service systemctl start rpcbind.service
检查 rpcinfo -p exportfs -r
其他node上执行： systemctl start rpcbind.service
测试： showmount -e 192.168.11.12
挂载磁盘 mount -t nfs 192.168.11.12:/data/k8s /data/k8s
安装yaml文件 kubectl create -f prometheus.configmap.yaml kubectl get configmaps -n kube-system |grep prometheus
注： configmap用于给容器内应用程序传递参数
kubectl create -f prometheus-volume.yaml kubectl get pvc –all-namespaces kubectl get pv prometheus
kubectl create -f prometheus-rbac.yaml
kubectl create -f prometheus.deploy.yaml kubectl get pod -n kube-system |grep prometheus
kubectl create -f prometeheus-svc.yaml
kubectl get svc -n kube-system |grep prometheus
prometheus NodePort 10.100.8.125
访问prometheus http://192.168.11.12:30378
kubectl create -f grafana.yaml
kubectl get svc -n kube-system |grep grafana
grafana-service NodePort 10.100.106.146
访问grafana http://192.168.11.12:32583
prometheus 监控 redis
kubectl create namespace abcdocker
kubectl create -f prometheus-redis-exporter.yaml
kubectl get pod -n abcdocker
kubectl get svc -n abcdocker
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
redis ClusterIP 10.102.102.242
curl 10.102.102.242:9121/metrics
kubectl delete -f prometheus.configmap.yaml
kubectl create -f prometheus.configmap.yaml
kubectl get svc -n kube-system |grep prometheus
prometheus NodePort 10.99.137.193
curl -X POST http://10.99.137.193:9090/-/reload
https://i4t.com/4134.html
node-exporter
kubectl create -f prometheus-node-exporter.yaml
| kubectl get pod -n kube-system -o wide | grep node |
任意节点测试： curl 127.0.0.1:9100/metrics
kubectl delete -f prometheus.configmap.yaml
kubectl create -f prometheus.configmap.yaml
kubectl get svc -n kube-system |grep prometheus
prometheus NodePort 10.99.137.193
curl -X POST http://10.99.137.193:9090/-/reload
页面上会报错
修改配置 vi prometheus.configmap.yaml 末尾添加： - job_name: ‘kubernetes-node’ kubernetes_sd_configs: - role: node relabel_configs: - source_labels: [address] regex: ‘(.*):10250’ replacement: ‘${1}:9100’ target_label: address action: replace
kubectl delete -f prometheus.configmap.yaml
kubectl create -f prometheus.configmap.yaml
curl -X POST http://10.99.137.193:9090/-/reload
https://i4t.com/4136.html
grafana 配置
https://i4t.com/4146.html
|
|