Developing a Go application with the Gin framework? Yep, it feels amazing to get that sleek, high-performance API running smoothly. But here’s the kicker: keeping an eye on the performance and spotting issues before they become nightmares is critical. Meet Prometheus, your soon-to-be favorite monitoring system. Let’s walk through setting it up with Gin, keeping things light and casual.

Getting the Gin Application Going

So, the first step, naturally, is to put together a basic Gin application. Picture this: You’ve got a blank canvas, your new Go module, and a main.go file waiting. Let’s give it some life with this:

package main

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

func main() {
    router := gin.Default()
    router.GET("/", func(context *gin.Context) {
        context.JSON(http.StatusOK, gin.H{
            "status": "ok",
        })
    })
    router.Run(":8080")
}

Boom! That’s your simple Gin server ready to tell the world it’s “status: ok” whenever you hit up the root URL.

Enter Prometheus

Alright, now for the star of the show: Prometheus. This tool is like magic when it comes to monitoring, and integrating it into your Gin app is a breeze. We’re using gin-prometheus-middleware from Carousell – it’s pretty much a fan favorite.

Middleware Installation

Let’s get this show on the road with a quick install:

go get github.com/carousell/gin-prometheus-middleware

Bringing in the Middleware

Next up, integrate this middleware into your application:

package main

import (
    "github.com/carousell/gin-prometheus-middleware"
    "github.com/gin-gonic/gin"
    "net/http"
)

func main() {
    r := gin.New()
    p := gpmiddleware.NewPrometheus("gin")
    p.Use(r)

    r.GET("/", func(c *gin.Context) {
        c.JSON(200, "Hello world! Visit /metrics for metrics")
    })

    r.Run(":37321")
}

Here, gpmiddleware.NewPrometheus("gin") creates the Prometheus magic, and p.Use(r) plugs it into your router. Now, hitting up /metrics shares a wealth of metrics just for the asking.

Metrics Galore

When you swing by /metrics, it’s like peeking into a treasure chest of data. You’ll find gems like:

• request_duration: Time spent handling requests.
• request_count: Total requests so far.
• request_total: Requests served, categorized by status code and method.

Going Custom

Sometimes, you need a little extra. Like, maybe you want to know how many operations your app has processed. Time to create custom metrics.

Custom Metrics Registration

Let’s add a counter for processed operations:

package main

import (
    "github.com/carousell/gin-prometheus-middleware"
    "github.com/gin-gonic/gin"
    "github.com/prometheus/client_golang/prometheus"
    "github.com/prometheus/client_golang/prometheus/promauto"
    "net/http"
    "time"
)

var (
    opsProcessed = promauto.NewCounter(prometheus.CounterOpts{
        Name: "myapp_processed_ops_total",
        Help: "The total number of processed operations",
    })
)

func recordMetrics() {
    go func() {
        for {
            opsProcessed.Inc()
            time.Sleep(2 * time.Second)
        }
    }()
}

func main() {
    r := gin.New()
    p := gpmiddleware.NewPrometheus("gin")
    p.Use(r)

    recordMetrics()

    r.GET("/", func(c *gin.Context) {
        c.JSON(200, "Hello world! Visit /metrics for metrics")
    })

    r.Run(":37321")
}

This snippet sets up a counter that ticks up every couple of seconds, giving you a running tally of processed operations.

Metrics Endpoint Setup

To make sure Prometheus can grab these juicy metrics, we need to handle the /metrics endpoint properly:

package main

import (
    "github.com/carousell/gin-prometheus-middleware"
    "github.com/gin-gonic/gin"
    "github.com/prometheus/client_golang/prometheus/promhttp"
    "net/http"
)

func main() {
    r := gin.New()
    p := gpmiddleware.NewPrometheus("gin")
    p.Use(r)

    r.GET("/metrics", gin.WrapH(promhttp.Handler()))

    r.GET("/", func(c *gin.Context) {
        c.JSON(200, "Hello world! Visit /metrics for metrics")
    })

    r.Run(":37321")
}

With promhttp.Handler(), you’re all set up. /metrics will now churn out all the monitoring data like a pro.

Up and Running

It’s showtime! Run that Go file with:

go run main.go

And then check out your metrics at http://localhost:37321/metrics. You can also roll with curl if you’re feeling command-line savvy:

curl http://localhost:37321/metrics

Boom, instant glance at how your app is performing.

Configuring Prometheus to Scrape

To reel all this data into a Prometheus server, a little configuration action is required. Let’s look at an example prometheus.yml:

scrape_configs:
  - job_name: myapp
    scrape_interval: 10s
    static_configs:
      - targets: ["localhost:37321"]

This script tells Prometheus to check in every 10 seconds and scrape the metrics from your app.

Wrapping Up

That’s it! Getting Prometheus integrated with your Gin application is super straightforward and incredibly useful. With middleware like gin-prometheus-middleware and the power to define custom metrics, you have everything you need to keep a close watch on your app. This setup isn’t just about avoiding pitfalls—it’s about truly understanding how your app ticks and making well-informed decisions to keep it humming smoothly.

Keep experimenting, keep monitoring, and watch as your insights help your Go applications rise to new heights of performance and reliability. Happy coding!