DEV Community

Cover image for Redis Queue and Cron in Go
fajar sp
fajar sp

Posted on

4

Redis Queue and Cron in Go

The original post is here

In this tutorial, we will interact with a queue and put it to a Redis server
using the github.com/hibiken/asynq package and create a scheduler for a
scheduled task using the github.com/robfig/cron package. This step-by-step
guide explains how to set up a queue, schedule tasks, and handle graceful
shutdowns.

Initialize the Module

Start by creating a new Go module for the project:

go mod init learn_queue_and_cron
Enter fullscreen mode Exit fullscreen mode

Create cron.go

The cron.go file is responsible for scheduling and running tasks at specific
intervals. Below is the implementation:

package main

import (
    "fmt"
    "log"
    "time"

    "github.com/robfig/cron/v3"
)

func runCron(c *cron.Cron) {

    // Schedule a task to run every minute
    _, err := c.AddFunc("@every 1m", func() {
        fmt.Printf("Task executed every minute at: %v \n", time.Now().Local())
    })
    if err != nil {
        log.Fatal(err)
    }

    // Start the cron scheduler
    c.Start()
    log.Println("Cron scheduler started")

    // Keep the main goroutine running
    select {}
}
Enter fullscreen mode Exit fullscreen mode

This code schedules a task to run every minute and keeps the application running
to ensure the scheduler works continuously.

Create queue.go

The queue.go file manages task processing using Asynq. Here's the code:

package main

import (
    "context"
    "encoding/json"
    "fmt"
    "log"

    "github.com/hibiken/asynq"
)

func runQueue(server *asynq.Server) {
    mux := asynq.NewServeMux()
    mux.HandleFunc("send_email", emailHandler)
    mux.HandleFunc("generate_report", reportHandler)

    if err := server.Run(mux); err != nil {
        log.Fatalf("Failed to run Asynq server: %v", err)
    }
}

func emailHandler(ctx context.Context, task *asynq.Task) error {
    var payload struct {
        To string `json:"to"`
    }
    if err := json.Unmarshal(task.Payload(), &payload); err != nil {
        return fmt.Errorf("failed to unmarshal payload: %w", err)
    }
    fmt.Printf("Sending email to: %s\n", payload.To)
    return nil
}

func reportHandler(ctx context.Context, task *asynq.Task) error {
    var payload struct {
        ReportID int `json:"report_id"`
    }
    if err := json.Unmarshal(task.Payload(), &payload); err != nil {
        return fmt.Errorf("failed to unmarshal payload: %w", err)
    }
    fmt.Printf("Generating report for ID: %d\n", payload.ReportID)
    return nil
}
Enter fullscreen mode Exit fullscreen mode

Explanation

  • Handlers: emailHandler and reportHandler process tasks by parsing their payloads and executing the respective actions.
  • Task Queue: Tasks such as "send_email" and "generate_report" are defined and processed via Asynq's task queue.

Create router.go

The router.go file sets up HTTP endpoints to enqueue tasks:

package main

import (
    "encoding/json"
    "net/http"

    "github.com/gin-gonic/gin"
    "github.com/hibiken/asynq"
)

func setupRouter(client *asynq.Client) *gin.Engine {
    r := gin.Default()

    r.POST("/enqueue/email", func(c *gin.Context) {
        var payload struct {
            To string `json:"to"`
        }
        if err := c.ShouldBindJSON(&payload); err != nil {
            c.JSON(http.StatusBadRequest, gin.H{"error": "Invalid request body"})
            return
        }

        jsonPayload, err := json.Marshal(payload)
        if err != nil {
            c.JSON(http.StatusInternalServerError, gin.H{"error": "Failed to marshal payload"})
            return
        }

        task := asynq.NewTask("send_email", jsonPayload)
        _, err = client.Enqueue(task)
        if err != nil {
            c.JSON(http.StatusInternalServerError, gin.H{"error": "Failed to enqueue task"})
            return
        }

        c.JSON(http.StatusOK, gin.H{"message": "Email job enqueued"})
    })

    r.POST("/enqueue/report", func(c *gin.Context) {
        var payload struct {
            ReportID int `json:"report_id"`
        }
        if err := c.ShouldBindJSON(&payload); err != nil {
            c.JSON(http.StatusBadRequest, gin.H{"error": "Invalid request body"})
            return
        }

        jsonPayload, err := json.Marshal(payload)
        if err != nil {
            c.JSON(http.StatusInternalServerError, gin.H{"error": "Failed to marshal payload"})
            return
        }

        task := asynq.NewTask("generate_report", jsonPayload)
        _, err = client.Enqueue(task)
        if err != nil {
            c.JSON(http.StatusInternalServerError, gin.H{"error": "Failed to enqueue task"})
            return
        }

        c.JSON(http.StatusOK, gin.H{"message": "Report job enqueued"})
    })

    return r
}
Enter fullscreen mode Exit fullscreen mode

This code uses the Gin framework to expose two endpoints for enqueuing tasks.

Create main.go

The main.go file integrates everything together:

package main

import (
    "context"
    "log"
    "net/http"
    "os"
    "os/signal"
    "syscall"
    "time"

    "github.com/hibiken/asynq"
    "github.com/robfig/cron/v3"
)

func main() {
    c := cron.New()

    server := asynq.NewServer(
        asynq.RedisClientOpt{Addr: "localhost:6379"},
        asynq.Config{
            Concurrency: 10,
        },
    )

    client := asynq.NewClient(asynq.RedisClientOpt{Addr: "localhost:6379"})
    defer client.Close()

    router := setupRouter(client)

    httpServer := &http.Server{
        Addr:    ":8080",
        Handler: router,
    }

    // Prepare shutdown context
    ctx, stop := context.WithCancel(context.Background())
    defer stop()
    quit := make(chan os.Signal, 1)
    signal.Notify(quit, os.Interrupt, syscall.SIGTERM)

    go runQueue(server)
    go runCron(c)
    go func() {
        if err := httpServer.ListenAndServe(); err != nil && err != http.ErrServerClosed {
            log.Fatalf("Failed to run HTTP server: %v", err)
        }
    }()

    appShutdown(ctx, httpServer, c, server, quit)
}

func appShutdown(ctx context.Context, httpServer *http.Server, c *cron.Cron, server *asynq.Server, quit chan os.Signal) {
    // Wait for termination signal
    <-quit
    log.Println("Shutting down gracefully...")

    httpCtx, httpCancel := context.WithTimeout(ctx, 5*time.Second)
    defer httpCancel()
    if err := httpServer.Shutdown(httpCtx); err != nil {
        log.Printf("HTTP server shutdown error: %v", err)
    }

    server.Shutdown()
    c.Stop()

    log.Println("Application stopped")
}
Enter fullscreen mode Exit fullscreen mode

This file combines the queue, cron, HTTP server, and shutdown logic.

Install Dependencies

Install all required dependencies:

go mod tidy
Enter fullscreen mode Exit fullscreen mode

Build and Run the Application

Build and run the application using:

go build -o run *.go && ./run
Enter fullscreen mode Exit fullscreen mode

Test the Application

Visit the following endpoints to enqueue tasks:

Watch the terminal for the task execution logs.

Canonical URL

For more detailed information, visit the original post on my blog.

ACI image

ACI.dev: Best Open-Source Composio Alternative (AI Agent Tooling)

100% open-source tool-use platform (backend, dev portal, integration library, SDK/MCP) that connects your AI agents to 600+ tools with multi-tenant auth, granular permissions, and access through direct function calling or a unified MCP server.

Star our GitHub!

Top comments (0)

ACI image

ACI.dev: Best Open-Source Composio Alternative (AI Agent Tooling)

100% open-source tool-use platform (backend, dev portal, integration library, SDK/MCP) that connects your AI agents to 600+ tools with multi-tenant auth, granular permissions, and access through direct function calling or a unified MCP server.

Star our GitHub!