Awesome

opentelemetry (This is a community driven project)

English | 中文

Opentelemetry for Hertz

Feature

Provider

• Out-of-the-box default opentelemetry provider
• Support setting via environment variables

Instrumentation

Tracing

• Support server and client hertz http tracing
• Support automatic transparent transmission of peer service through http headers

Metrics

• Support hertz http metrics [R.E.D]
• Support service topology map metrics [Service Topology Map]
• Support go runtime metrics

Logging

• Extend hertz logger based on logrus
• Implement tracing auto associated logs

Configuration via environment variables

• Exporter
• SDK

Server usage

import (
    // ...
    "github.com/hertz-contrib/obs-opentelemetry/provider"
    hertztracing "github.com/hertz-contrib/obs-opentelemetry/tracing"
)


func main()  {
    serviceName := "echo"

    p := provider.NewOpenTelemetryProvider(
        provider.WithServiceName(serviceName),
        provider.WithExportEndpoint("localhost:4317"),
        provider.WithInsecure(),
    )
    defer p.Shutdown(context.Background())

    tracer, cfg := hertztracing.NewServerTracer()
    h := server.Default(tracer)
    h.Use(hertztracing.ServerMiddleware(cfg))

    // ...

    h.Spin()
}

Client usage

import (
    // ...
    "github.com/hertz-contrib/obs-opentelemetry/provider"
    hertztracing "github.com/hertz-contrib/obs-opentelemetry/tracing"
)

func main(){
    serviceName := "echo-client"

    p := provider.NewOpenTelemetryProvider(
        provider.WithServiceName(serviceName),
        provider.WithExportEndpoint("localhost:4317"),
        provider.WithInsecure(),
    )
    defer p.Shutdown(context.Background())

    c, _ := client.NewClient()
    c.Use(hertztracing.ClientMiddleware())

    // ...

}

Tracing associated Logs

set logger impl

import (
    hertzlogrus "github.com/hertz-contrib/obs-opentelemetry/logging/logrus"
)

func init()  {
    hlog.SetLogger(hertzlogrus.NewLogger())
    hlog.SetLevel(hlog.LevelDebug)

}

You need to configure the tracing middleware in the above example at the same time. If you do not need to report the tracing, you can refer to issue

log with context

h.GET("/ping", func(c context.Context, ctx *app.RequestContext) {
    req := &api.Request{Message: "my request"}
    resp, err := client.Echo(c, req)
    if err != nil {
        hlog.Errorf(err.Error())
    }
    hlog.CtxDebugf(c, "message received successfully: %s", req.Message)
    ctx.JSON(consts.StatusOK, resp)
})

view log

{"level":"debug","msg":"message received successfully: my request","span_id":"445ef16484a171b8","time":"2022-07-04T06:27:35+08:00","trace_flags":"01","trace_id":"e9e579b32c9d6b0598f8f33d65689e06"}

Example

Executable Example

Supported Metrics

HTTP Request Metrics

Hertz Server

Below is a table of HTTP server metric instruments.

Hertz Client

Below is a table of HTTP client metric instruments.

R.E.D

The RED Method defines the three key metrics you should measure for every microservice in your architecture. We can calculate RED based on http.server.duration.

Rate

the number of requests, per second, you services are serving.

eg: QPS

sum(rate(http_server_request_count_total{}[5m])) by (service_name, http_method)

Errors

the number of failed requests per second.

eg: Error ratio

sum(rate(http_server_request_count_total{status_code="Error"}[5m])) by (service_name, http_method) / sum(rate(http_server_request_count_total{}[5m])) by (service_name, http_method)

Duration

distributions of the amount of time each request takes

eg: P99 Latency

histogram_quantile(0.99, sum(rate(http_server_duration_bucket{}[5m])) by (le, service_name, http_method))

Service Topology Map

The http.server.duration will record the peer service and the current service dimension. Based on this dimension, we can aggregate the service topology map

sum(rate(http_server_request_count_total{}[5m])) by (service_name, peer_service)

Runtime Metrics

Compatibility

The sdk of OpenTelemetry is fully compatible with 1.X opentelemetry-go. see

maintained by: CoderPoet

Dependencies