１ 迁移需要的数据表
配置
修改.env文件,队列支持"sync", "database", "beanstalkd", "sqs", "redis", "null",这里采用数据库,其他处理方式也都一样。
# mysql数据库
DB_CONNECTION=mysql
DB_HOST=127.0.0.1
DB_PORT=3306
DB_DATABASE=test
DB_USERNAME=root
DB_PASSWORD=root
# 队列驱动
QUEUE_DRIVER=database
执行
# 创建队列作业的表
php artisan queue:table
# 迁移数据
php artisan migrate
２ 创建任务类
php artisan make:job SendEmail
生成文件app/Jobs/SendEmail.php
<?php
namespace App\Jobs;
use Illuminate\Bus\Queueable;
use Illuminate\Queue\SerializesModels;
use Illuminate\Queue\InteractsWithQueue;
use Illuminate\Contracts\Queue\ShouldQueue;
use Illuminate\Foundation\Bus\Dispatchable;
class SendEmail implements ShouldQueue
{
use Dispatchable, InteractsWithQueue, Queueable, SerializesModels;
/**
* Create a new job instance.
*
* @return void
*/
public function __construct()
{
//
}
/**
* Execute the job.
*
* @return void
*/
public function handle()
{
//
}
}
handle 方法为队列处理的具体逻辑,我们这里添加一行日志记录
<?php
namespace App\Jobs;
use Illuminate\Bus\Queueable;
use Illuminate\Queue\SerializesModels;
use Illuminate\Queue\InteractsWithQueue;
use Illuminate\Contracts\Queue\ShouldQueue;
use Illuminate\Foundation\Bus\Dispatchable;
class SendEmail implements ShouldQueue
{
use Dispatchable, InteractsWithQueue, Queueable, SerializesModels;
protected $email;
/**
* Create a new job instance.
*
* @return void
*/
public function __construct($email)
{
$this->email=$email;
}
/**
* Execute the job.
*
* @return void
*/
public function handle()
{
\Log::info("发送邮件到 -> ".$this->email);
}
}
３ 推送任务
Controller
<?php
namespace App\Http\Controllers;
use App\Jobs\SendEmail;
use Illuminate\Foundation\Bus\DispatchesJobs;
use Illuminate\Routing\Controller as BaseController;
use Illuminate\Foundation\Validation\ValidatesRequests;
use Illuminate\Foundation\Auth\Access\AuthorizesRequests;
class Controller extends BaseController
{
use AuthorizesRequests, DispatchesJobs, ValidatesRequests;
public function index(){
dispatch(new SendEmail('xxxx@demo.com'));
$this->dispatch(new SendEmail('yyyy@demo.com'));
}
}
route
<?php
/*
|--------------------------------------------------------------------------
| Web Routes
|--------------------------------------------------------------------------
|
| Here is where you can register web routes for your application. These
| routes are loaded by the RouteServiceProvider within a group which
| contains the "web" middleware group. Now create something great!
|
*/
Route::get('/', function () {
return view('welcome');
});
Route::get('/index','Controller@index');
推送
curl http://localhost:8084/index
查看jobs表
image.png
４ 执行任务
$ php artisan queue:listen
[2019-03-15 04:39:24] Processing: App\Jobs\SendEmail
[2019-03-15 04:39:24] Processed: App\Jobs\SendEmail
[2019-03-15 04:39:24] Processing: App\Jobs\SendEmail
[2019-03-15 04:39:24] Processed: App\Jobs\SendEmail
查看jobs表,发现任务已经执行完成
image.png
指定最大尝试次数和超时时间
$ php artisan queue:listen --tries=3 --timeout=30``
也可以写到代码里
<?php
namespace App\Jobs;
use Illuminate\Bus\Queueable;
use Illuminate\Queue\SerializesModels;
use Illuminate\Queue\InteractsWithQueue;
use Illuminate\Contracts\Queue\ShouldQueue;
use Illuminate\Foundation\Bus\Dispatchable;
class SendEmail implements ShouldQueue
{
use Dispatchable, InteractsWithQueue, Queueable, SerializesModels;
/**
* 任务最大尝试次数。
*
* @var int
*/
public $tries = 3;
/**
* 任务运行的超时时间。
*
* @var int
*/
public $timeout = 120;
protected $email;
/**
* Create a new job instance.
*
* @return void
*/
public function __construct($email)
{
$this->email=$email;
}
/**
* Execute the job.
*
* @return void
*/
public function handle()
{
//throw new \Exception("发生异常");
/*
\Mail::raw("队列测试",function ($message){
$message->to($this->email);
});
*/
\Log::info("发送邮件到 -> ".$this->email);
}
}
5 处理失败任务
有时候你队列中的任务会失败。不要担心，本来事情就不会一帆风顺。Laravel 内置了一个方便的方式来指定任务重试的最大次数。当任务超出这个重试次数后，它就会被插入到 failed_jobs 数据表里面。要创建 failed_jobs 表的迁移文件，你可以用 queue:failed-table 命令，接着使用 migrate Artisan 命令生成 failed_jobs 表
# 创建队列作业的表
php artisan queue:failed-table
# 迁移数据
php artisan migrate
查看所有失败任务 php artisan queue:failed
- ID 任务的 ID (任务 ID 可以被用在重试失败的任务上)
- Connection 连接
- Queue 队列
- Failed At 失败时间
image.png
6 失败任务重试
重试ID 为 24的失败任务,此时会将任务重新放入队列,生成新的ID
$ php artisan queue:retry 24
The failed job [24] has been pushed back onto the queue!
重试所有失败的任务使用 all 作为 ID：
$ php artisan queue:retry all
The failed job [23] has been pushed back onto the queue!
The failed job [22] has been pushed back onto the queue!
The failed job [21] has been pushed back onto the queue!
The failed job [20] has been pushed back onto the queue!
The failed job [19] has been pushed back onto the queue!
The failed job [18] has been pushed back onto the queue!
The failed job [17] has been pushed back onto the queue!
The failed job [16] has been pushed back onto the queue!
The failed job [15] has been pushed back onto the queue!
The failed job [14] has been pushed back onto the queue!
The failed job [13] has been pushed back onto the queue!
The failed job [12] has been pushed back onto the queue!
The failed job [11] has been pushed back onto the queue!
The failed job [10] has been pushed back onto the queue!
The failed job [9] has been pushed back onto the queue!
The failed job [8] has been pushed back onto the queue!
The failed job [7] has been pushed back onto the queue!
The failed job [6] has been pushed back onto the queue!
The failed job [5] has been pushed back onto the queue!
The failed job [4] has been pushed back onto the queue!
The failed job [3] has been pushed back onto the queue!
The failed job [2] has been pushed back onto the queue!
The failed job [1] has been pushed back onto the queue!
6 失败任务删除
删除ID为74的任务
php artisan queue:forget 74
image.png
删除所有失败的任务
php artisan queue:flush
image.png
命令
queue:failed 列出所有失败的队列
queue:failed-table 创建失败队列数据表迁移
queue:flush 清除所有失败的队列作业
queue:forget 删除失败的队列作业
queue:listen 监听队列
queue:restart 重新启动队列工作进程
queue:retry 重试失败的队列
queue:table 创建队列数据库表迁移
queue:work 作为守护程序开始处理队列上的作业
queue:listen和queue:work 区别
- queue:work - 这是新的“守护进程”进程（不再需要该标志）。该框架将启动“一次” - 然后继续循环工作。这将无限期地继续下去。它使用较少的内存/ CPU，queue:listen因为框架在整个时间内保持不变。您还必须记住使用queue:restart强制队列来更新在修补期间推送的任何代码更改。
- queue:work --once - 这将激活框架，处理一个作业，然后关闭。适用于开发过程中的测试等
- queue:listen - 这将在每个周期启动框架，处理一个作业，然后完全关闭，然后再次启动框架等，并无限循环。这意味着在处理每个作业后释放所有内存/进程。如果你有内存泄漏queue:work- 尝试一下。
