在 Laravel 中实现可视化监控

在 Laravel 中实现可视化监控，可以通过以下 5 层架构构建完整的监控体系：

架构设计

一、数据采集层

1. HTTP 请求监控

创建全局中间件记录请求指标：

// app/Http/Middleware/MonitorRequests.php
public function handle(Request $request, Closure $next)
{
    $start = microtime(true);
    
    $response = $next($request);
    
    $duration = microtime(true) - $start;
    
    DB::table('request_metrics')->insert([
        'method' => $request->method(),
        'path' => $request->path(),
        'status_code' => $response->getStatusCode(),
        'duration' => $duration,
        'ip' => $request->ip(),
        'created_at' => now()
    ]);
    
    return $response;
}

2. 异常监控

在异常处理类中记录：

// app/Exceptions/Handler.php
public function report(Throwable $e)
{
    DB::table('error_logs')->insert([
        'message' => $e->getMessage(),
        'stack_trace' => $e->getTraceAsString(),
        'file' => $e->getFile(),
        'line' => $e->getLine(),
        'created_at' => now()
    ]);
    
    parent::report($e);
}

二、数据存储层

1. 时序数据表结构

Schema::create('system_metrics', function (Blueprint $table) {
    $table->id();
    $table->string('metric_name'); // CPU/Memory/QueueSize
    $table->double('value');
    $table->timestamp('timestamp');
});

Schema::create('business_metrics', function (Blueprint $table) {
    $table->id();
    $table->string('type'); // sms_sent/sms_failed/payment
    $table->integer('count')->default(0);
    $table->date('date');
});

2. 定时采集系统指标

创建 Artisan 命令：

// app/Console/Commands/CollectSystemMetrics.php
public function handle()
{
    $cpuUsage = sys_getloadavg()[0];
    $memoryUsage = memory_get_usage(true) / 1024 / 1024;

    DB::table('system_metrics')->insert([
        ['metric_name' => 'cpu', 'value' => $cpuUsage, 'timestamp' => now()],
        ['metric_name' => 'memory', 'value' => $memoryUsage, 'timestamp' => now()]
    ]);
}

三、可视化呈现层

方案1：使用 Grafana + Prometheus

1. 安装 Prometheus 客户端

comrequire promphp/prometheus_client_php

2. 暴露指标接口

// routes/metrics.php
Route::get('/metrics', function() {
    $registry = new CollectorRegistry(new InMemory());
    
    $counter = $registry->getOrRegisterCounter(
        'app', 
        'http_requests_total', 
        'Total HTTP requests',
        ['method', 'endpoint', 'status']
    );
    
    // 从数据库获取数据填充指标
    $requests = DB::table('request_metrics')->get();
    foreach ($requests as $request) {
        $counter->incBy(1, [
            $request->method,
            $request->path,
            $request->status_code
        ]);
    }
    
    return response($registry->getMetricFamilySamples())
           ->header('Content-Type', 'text/plain');
});

3. Grafana 仪表板配置

# prometheus.yml 配置
scrape_configs:
  - job_name: 'laravel'
    static_configs:
      - targets: ['laravel-app:9600']

方案2：使用内置视图

创建监控面板视图：

// resources/views/admin/metrics.blade.php
<div class="grid grid-cols-4 gap-4">
    <div class="bg-white p-4 shadow">
        <h3>实时请求频率</h3>
        <canvas id="requestsChart"></canvas>
    </div>
    
    <div class="bg-white p-4 shadow">
        <h3>系统资源</h3>
        <div id="cpuGauge"></div>
        <div id="memoryGauge"></div>
    </div>
</div>

@push('scripts')
<script>
// 使用 Chart.js 展示请求趋势
new Chart(document.getElementById('requestsChart'), {
    type: 'line',
    data: {
        labels: {!! $requestTimes->pluck('time') !!},
        datasets: [{
            label: '请求延迟(ms)',
            data: {!! $requestTimes->pluck('duration') !!}
        }]
    }
});
</script>
@endpush

四、报警通知层

1. 异常阈值检测

创建监控任务：

// app/Console/Commands/MonitorSystem.php
public function handle()
{
    // 检测错误率
    $errorCount = DB::table('error_logs')
        ->where('created_at', '>', now()->subMinutes(5))
        ->count();
    
    if ($errorCount > 20) {
        Notification::route('slack', config('services.slack.webhook'))
            ->notify(new HighErrorRate($errorCount));
    }
    
    // 检测队列积压
    $queueSize = DB::table('jobs')->count();
    if ($queueSize > 1000) {
        Mail::to('admin@example.com')->send(new QueueBacklogAlert($queueSize));
    }
}

2. 报警规则配置

// config/monitoring.php
return [
    'alerts' => [
        'error_rate' => [
            'threshold' => 20,
            'time_window' => 5, // 分钟
            'channels' => ['slack', 'email']
        ],
        'queue_backlog' => [
            'threshold' => 1000,
            'channels' => ['sms']
        ]
    ]
];

五、优化方案对比

扩展监控维度

1. 业务指标监控

// 在短信发送逻辑中添加
DB::table('business_metrics')->updateOrInsert(
    ['type' => 'sms_sent', 'date' => now()->format('Y-m-d')],
    ['count' => DB::raw('count + 1')]
);

2. 用户行为分析

// 记录用户操作事件
event(new UserAction(
    auth()->id(),
    'sent_sms',
    ['phone' => $request->phone]
));

3. 分布式追踪
   集成 OpenTelemetry：

comrequire open-telemetry/opentelemetry

通过以上方案可实现：

• ✅ 实时系统健康监控
• ✅ 历史数据分析
• ✅ 多维度报警机制
• ✅ 业务指标可视化
• ✅ 从基础设施到应用层的完整观测