当前位置：首页 > news >正文

可观测性：日志、指标与追踪

news 2026/5/30 22:55:42

可观测性：日志、指标与追踪

大家好，我是欧阳瑞（Rich Own）。今天想和大家聊聊可观测性这个重要话题。作为一个全栈开发者，可观测性是保障系统稳定运行的关键。今天就来分享一下日志、指标与追踪的实战经验。

可观测性概述

什么是可观测性？

可观测性是指通过外部输出（日志、指标、追踪）理解系统内部状态的能力

三大支柱

支柱	说明	工具
日志	事件的结构化记录	ELK、Loki
指标	数值型数据的聚合	Prometheus
追踪	分布式系统的请求路径	Jaeger、Zipkin

日志管理

日志收集

const winston = require('winston'); const logger = winston.createLogger({ level: 'info', format: winston.format.json(), transports: [ new winston.transports.File({ filename: 'error.log', level: 'error' }), new winston.transports.File({ filename: 'combined.log' }), new winston.transports.Console({ format: winston.format.simple() }) ] }); // 使用 logger.info('User logged in', { userId: '123' }); logger.error('Database connection failed', { error: err.message });

日志结构化

const structuredLogger = winston.createLogger({ format: winston.format.combine( winston.format.timestamp(), winston.format.errors({ stack: true }), winston.format.json() ), transports: [new winston.transports.Console()] }); // 结构化日志输出 // { // "level": "info", // "message": "Order created", // "timestamp": "2024-01-01T12:00:00.000Z", // "orderId": "abc123", // "userId": "456" // }

指标监控

Prometheus配置

# prometheus.yml global: scrape_interval: 15s scrape_configs: - job_name: 'node_exporter' static_configs: - targets: ['localhost:9100'] - job_name: 'app_metrics' static_configs: - targets: ['localhost:3000']

自定义指标

const client = require('prom-client'); const httpRequestDuration = new client.Histogram({ name: 'http_request_duration_seconds', help: 'Duration of HTTP requests in seconds', labelNames: ['method', 'route', 'status_code'] }); const requestCounter = new client.Counter({ name: 'http_requests_total', help: 'Total number of HTTP requests', labelNames: ['method', 'route', 'status_code'] }); // 在中间件中使用 app.use((req, res, next) => { const start = Date.now(); res.on('finish', () => { const duration = (Date.now() - start) / 1000; httpRequestDuration.observe({ method: req.method, route: req.path, status_code: res.statusCode }, duration); requestCounter.inc({ method: req.method, route: req.path, status_code: res.statusCode }); }); next(); }); // 暴露指标端点 app.get('/metrics', (req, res) => { res.set('Content-Type', client.register.contentType); res.send(client.register.metrics()); });

分布式追踪

OpenTelemetry配置

const { NodeSDK } = require('@opentelemetry/sdk-node'); const { ConsoleSpanExporter } = require('@opentelemetry/sdk-trace-base'); const { getNodeAutoInstrumentations } = require('@opentelemetry/auto-instrumentations-node'); const sdk = new NodeSDK({ traceExporter: new ConsoleSpanExporter(), instrumentations: [getNodeAutoInstrumentations()] }); sdk.start();

自定义追踪

const { trace } = require('@opentelemetry/api'); const tracer = trace.getTracer('my-service'); async function processOrder(orderId) { return tracer.startActiveSpan('process-order', async (span) => { span.setAttribute('orderId', orderId); try { await validateOrder(orderId); await updateInventory(orderId); await sendNotification(orderId); } finally { span.end(); } }); }

实战案例：完整监控系统

class MonitoringSystem { constructor() { this.logger = this.setupLogger(); this.metrics = this.setupMetrics(); this.tracer = this.setupTracer(); } setupLogger() { return winston.createLogger({ level: 'info', format: winston.format.json(), transports: [new winston.transports.Console()] }); } setupMetrics() { const client = require('prom-client'); return { requests: new client.Counter({ name: 'app_requests_total', help: 'Total requests' }), latency: new client.Histogram({ name: 'app_request_latency_seconds', help: 'Request latency' }) }; } setupTracer() { const { trace } = require('@opentelemetry/api'); return trace.getTracer('app-tracer'); } log(level, message, metadata) { this.logger[level](message, metadata); } recordRequest(duration) { this.metrics.requests.inc(); this.metrics.latency.observe(duration); } }

最佳实践

1. 统一日志格式

// 使用ECS格式 const ecsFormat = winston.format((info) => { return { '@timestamp': info.timestamp, 'log.level': info.level, 'message': info.message, 'service.name': 'my-service', ...info.metadata }; });

2. 设置告警规则

# Prometheus Alertmanager配置 groups: - name: example rules: - alert: HighErrorRate expr: sum(rate(http_requests_total{status_code="5xx"}[5m])) / sum(rate(http_requests_total[5m])) > 0.1 for: 5m labels: severity: critical annotations: summary: High error rate detected