一、前言
监控指标
Go监控指标有好多库，监控指标大致一样。
1）https://github.com/bmhatfield/go-runtime-metrics
2）https://github.com/tevjef/go-runtime-metrics
3）Prometheus SDK
指标输出
1）golang expvar
2）Prometheus
二、Prometheus搭建
镜像设置
docker pull mirrorgooglecontainers/addon-resizer:1.8.3
docker tag mirrorgooglecontainers/addon-resizer:1.8.3 k8s.gcr.io/addon-resizer:1.8.3
docker pull mirrorgooglecontainers/heapster-grafana-amd64:v5.0.4
docker tag mirrorgooglecontainers/heapster-grafana-amd64:v5.0.4 k8s.gcr.io/heapster-grafana-amd64:v5.0.4
prometheus部署
kubectl apply -f alertmanager
kubectl apply -f prometheus
kubectl apply -f kube-state-metrics
kubectl apply -f node-exporter
kubectl apply -f grafana
检查
kubectl get pod -n monitoring
kubectl get svc -n monitoring
三、Grafana搭建
数据源配置
image.png
Dashboard配置
https://grafana.com/grafana/dashboards/6671
四、效果展示
自定义job
image.png
查看指标
image.png
dashboard展示pod信息
image.png
实现
prometheus参考Thakurvaibhav的作品，并进行改造。
demo项目使用prometheus-client
代码见：
https://github.com/huiwq1990/OpenShare/tree/master/monitoring
参考
https://medium.com/faun/production-grade-kubernetes-monitoring-using-prometheus-78144b835b60
https://github.com/Thakurvaibhav/k8s/blob/master/monitoring/README.md
https://github.com/giantswarm/prometheus
https://o-my-chenjian.com/2018/01/27/Deploy-Prometheus-And-Grafana-On-Kubernetes/
