prometheus

Contents

1. 1. play - node info metrics

Prometheus collects and stores its metrics as time series data

Components

• Prometheus server: scrapes and stores time series data 单文件
• alertmanager
• exporter https://www.prometheus.wang/exporter/what-is-prometheus-exporter.html
  • node_exporter
• Grafana, for visualize [optional]

play - node info metrics

1. run exporter to generate metrics
2. add config to server to listen

   scrape_configs:
     - job_name:       'node'
   
       # Override the global default and scrape targets from this job every 5 seconds.
       scrape_interval: 5s
   
       static_configs:
         - targets: ['192.168.56.105:8081', '192.168.56.105:8082']
           labels:
             group: 'production'
   
         - targets: ['localhost:8082']
           labels:
             group: 'canary'

定义了4中不同的指标类型(metric type)：Counter（计数器）、Gauge（仪表盘）、Histogram（直方图）、Summary（摘要）。

Counter：只增不减的计数器
rate(http_requests_total[5m])
topk(10, http_requests_total)
Gauge：可增可减的仪表盘
node_memory_MemFree
delta(cpu_temp_celsius{host=”zeus”}[2h])