json datasource使用

概述

在grafana的数据源中，有个比较轻量级的数据源，可以跨网络进行json数据访问，比较灵活，对于需要由grafana来渲染数据，可以通过这种方式进行暴露。

安装

进入grafana服务器，执行以下命令进行安装，重启后即可使用json数据源

grafana-cli plugins install simpod-json-datasource

json数据源服务

这里，我使用了gin框架，实现了json数据源至少需要的3个访问点，1个用于测试连通性，其余2个为指标数据返回。

json数据源的使用官方参考链接：https://grafana.com/grafana/plugins/simpod-json-datasource/#development-setup

To work with this datasource the backend needs to implement 4 endpoints:

GET / with 200 status code response. Used for “Test connection” on the datasource config page.
POST /search to return available metrics.
POST /query to return panel data or annotations.

Those two urls are optional:

POST /tag-keys returning tag keys for ad hoc filters.
POST /tag-values returning tag values for ad hoc filters.

实现代码如下

package main

import (
   "fmt"
   "github.com/gin-gonic/gin"
   "net/http"
)

type Metric struct {
   Target     string          `json:"target"`
   Datapoints [][]interface{} `json:"datapoints"`
}

func newMetric(target string, data [][]interface{}) *Metric {
   return &Metric{
      Target:     target,
      Datapoints: data,
   }
}

func main() {
   r := gin.Default()

   //对于json datasource，需要实现3个访问点
   //1.GET / with 200 status code response. Used for "Test connection" on the datasource config page.
   r.GET("/", func(c *gin.Context) {
      c.JSON(200, gin.H{
         "message": "pong",
      })
   })

   //2.POST /search to return available metrics.
   r.POST("/search", func(context *gin.Context) {
      //查看grafana提交过来的body内容
      body, _ := context.GetRawData()
      fmt.Println("Body:", string(body))
      context.JSON(http.StatusOK, []interface{}{"demo", 22, "test", "share"})
   })

   //3.POST /query to return panel data or annotations.
   m1 := newMetric("pps in", [][]interface{}{
      {"ok", 1450754160000},
      {"error", 1450754220000},
   })
   m2 := newMetric("pps out", [][]interface{}{
      {861, 1450754160000},
      {767, 1450754220000},
   })
   m3 := newMetric("errors out", [][]interface{}{
      {861, 1450754160000},
      {767, 1450754220000},
   })
   m4 := newMetric("errors in", [][]interface{}{
      {861, 1450754160000},
      {767, 1450754220000},
   })
   result := []*Metric{
      m1, m2, m3, m4,
   }
   r.POST("/query", func(context *gin.Context) {
      //查看grafana提交过来的body内容
      body, _ := context.GetRawData()
      fmt.Println("Body:", string(body))
      context.JSON(http.StatusOK, result)
   })
   r.Run("0.0.0.0:58080") // listen and serve on 0.0.0.0:8080
}

测试结果

建立数据源并测试

此过程，grafana通过jsonDatasource访问目标的端点服务 /

服务端输出

[GIN] 2021/09/15 - 16:43:35 | 200 |     165.478?s |      172.17.0.1 | GET      "/"

获取指标项目

此过程，grafana通过jsonDatasource访问目标的端点服务 /search，并提交{“target”:""}的json数据

服务端输出

Body: {"target":""}
[GIN] 2021/09/15 - 16:45:38 | 200 |     207.425µs |      172.17.0.1 | POST     "/search"

获取指标数据

此过程，grafana通过jsonDatasource访问目标的端点服务 /query，并提交了json数据

服务端输出

Body: {"app":"dashboard","requestId":"Q174","timezone":"browser","panelId":123125,"dashboardId":null,"range":{"from":"2021-09-15T03:58:50.983Z","to":"2021-09-15T09:58:50.983Z","raw":{"from":"now-6h","to":"now"}},"timeInfo":"","interval":"15s","intervalMs":15000,"targets":[{"refId":"A","payload":"","target":"时序性数据","datasource":"JSON"}],"maxDataPoints":1318,"scopedVars":{"__interval":{"text":"15s","value":"15s"},"__interval_ms":{"text":"15000","value":15000}},"startTime":1631699930983,"rangeRaw":{"from":"now-6h","to":"now"},"adhocFilters":[]}
[GIN] 2021/09/15 - 16:50:26 | 200 |     265.951µs |      172.17.0.1 | POST     "/query"

从获取的json数据可以看到，我们可以通过json数据要求，传送对应的时序数据给grafana进行数据展示，如果默认的数据请求不够，可以通过payload进行内容限定。

Body: {"app":"dashboard","requestId":"Q156","timezone":"browser","panelId":123125,"dashboardId":null,"range":{"from":"2021-09-15T02:50:51.665Z","to":"2021-09-15T08:50:51.665Z","raw":{"from":"now-6h","to":"now"}},"timeInfo":"","interval":"30s","intervalMs":30000,"targets":[{"refId":"A","payload":{"require":"demo"},"target":22,"datasource":"JSON"}],"maxDataPoints":581,"scopedVars":{"__interval":{"text":"30s","value":"30s"},"__interval_ms":{"text":"30000","value":30000}},"startTime":1631696138287,"rangeRaw":{"from":"now-6h","to":"now"},"adhocFilters":[]}
[GIN] 2021/09/15 - 16:55:39 | 200 |     205.214µs |      172.17.0.1 | POST     "/query"

获取表格数据

总结

从以上例子可以看到，我们可以非常轻便地通过web服务端点，暴露出希望由grafana进行渲染的数据内容，增强数据的可视化
grafana展示的数据大部分是以时序数据为主的，json datasource的数据可以具有时序的一列，并且是 unix timestamp in milliseconds
json datasource可以展示表格数据，只要满足要求的json数据，就可以进行渲染
json datasource仅支持数据展示，无法进行操作反馈

json datasource使用

Go语言相关栏目本月热门文章