Node.js

在不到 5 分钟内为您的应用获取遥测数据!

本页面将向您展示如何在 Node.js 中开始使用 OpenTelemetry。

您将学习如何 instrument tracesmetrics,并将它们记录到控制台。

先决条件

请确保您已在本地安装以下软件

示例应用程序

以下示例使用了一个基本的 Express 应用程序。如果您不使用 Express,也没关系 — 您也可以将 OpenTelemetry JavaScript 与其他 Web 框架(如 Koa 和 Nest.JS)一起使用。有关支持框架的库的完整列表,请参阅 注册表

有关更详细的示例,请参阅 示例

依赖项

首先,在一个新目录中设置一个空的 package.json 文件。

npm init -y

接下来,安装 Express 依赖项。

npm install express @types/express
npm install -D tsx  # a tool to run TypeScript (.ts) files directly with node
npm install express

创建并启动 HTTP 服务器

创建一个名为 app.ts 的文件(如果未使用 TypeScript,则为 app.js)并添加以下代码:

/*app.ts*/
import express, { Express } from 'express';

const PORT: number = parseInt(process.env.PORT || '8080');
const app: Express = express();

function getRandomNumber(min: number, max: number) {
  return Math.floor(Math.random() * (max - min + 1) + min);
}

app.get('/rolldice', (req, res) => {
  res.send(getRandomNumber(1, 6).toString());
});

app.listen(PORT, () => {
  console.log(`Listening for requests on https://:${PORT}`);
});
/*app.js*/
const express = require('express');

const PORT = parseInt(process.env.PORT || '8080');
const app = express();

function getRandomNumber(min, max) {
  return Math.floor(Math.random() * (max - min + 1) + min);
}

app.get('/rolldice', (req, res) => {
  res.send(getRandomNumber(1, 6).toString());
});

app.listen(PORT, () => {
  console.log(`Listening for requests on https://:${PORT}`);
});

使用以下命令运行应用程序,然后在浏览器中打开 https://:8080/rolldice 以确保其正常工作。

$ npx tsx app.ts
Listening for requests on https://:8080
$ node app.js
Listening for requests on https://:8080

仪表

以下内容展示了如何安装、初始化和运行一个已 instrumented 的 OpenTelemetry 应用程序。

更多依赖项

首先,安装 Node SDK 和 autoinstrumentations 包。

Node SDK 允许您使用对大多数用例都适用的默认配置来初始化 OpenTelemetry。

auto-instrumentations-node 包安装了 instrumentation 库,这些库将自动创建与库中调用的代码相对应的 spans。在这种情况下,它为 Express 提供了 instrumentation,使示例应用程序能够自动为每个传入的请求创建 spans。

npm install @opentelemetry/sdk-node \
  @opentelemetry/api \
  @opentelemetry/auto-instrumentations-node \
  @opentelemetry/sdk-metrics \
  @opentelemetry/sdk-trace-node

要查找所有 autoinstrumentation 模块,您可以查看 注册表

设置

Instrumentation 的设置和配置必须在您的应用程序代码运行之前执行。一个常用的工具是 –import 标志。

创建一个名为 instrumentation.ts 的文件(如果未使用 TypeScript,则为 instrumentation.mjs),其中将包含您的 instrumentation 设置代码。

/*instrumentation.ts*/
import { NodeSDK } from '@opentelemetry/sdk-node';
import { ConsoleSpanExporter } from '@opentelemetry/sdk-trace-node';
import { getNodeAutoInstrumentations } from '@opentelemetry/auto-instrumentations-node';
import {
  PeriodicExportingMetricReader,
  ConsoleMetricExporter,
} from '@opentelemetry/sdk-metrics';

const sdk = new NodeSDK({
  traceExporter: new ConsoleSpanExporter(),
  metricReader: new PeriodicExportingMetricReader({
    exporter: new ConsoleMetricExporter(),
  }),
  instrumentations: [getNodeAutoInstrumentations()],
});

sdk.start();
/*instrumentation.mjs*/
import { NodeSDK } from '@opentelemetry/sdk-node';
import { ConsoleSpanExporter } from '@opentelemetry/sdk-trace-node';
import { getNodeAutoInstrumentations } from '@opentelemetry/auto-instrumentations-node';
import {
  PeriodicExportingMetricReader,
  ConsoleMetricExporter,
} from '@opentelemetry/sdk-metrics';

const sdk = new NodeSDK({
  traceExporter: new ConsoleSpanExporter(),
  metricReader: new PeriodicExportingMetricReader({
    exporter: new ConsoleMetricExporter(),
  }),
  instrumentations: [getNodeAutoInstrumentations()],
});

sdk.start();

运行已 instrumented 的应用

现在您可以像往常一样运行您的应用程序,但可以使用 --import 标志在应用程序代码加载之前加载 instrumentation。确保您的 NODE_OPTIONS 环境变量中没有其他冲突的 --import--require 标志,例如 --require @opentelemetry/auto-instrumentations-node/register

$ npx tsx --import ./instrumentation.ts app.ts
Listening for requests on https://:8080
$ node --import ./instrumentation.mjs app.js
Listening for requests on https://:8080

在浏览器中打开 https://:8080/rolldice 并刷新几次页面。过一会儿,您应该会在控制台中看到由 ConsoleSpanExporter 打印的 spans。

查看示例输出
{
  resource: {
    attributes: {
      'host.arch': 'arm64',
      'host.id': '8FEBBC33-D6DA-57FC-8EF0-1A9C14B919F8',
      'process.pid': 12460,
      // ... some resource attributes elided ...
      'process.runtime.version': '22.17.1',
      'process.runtime.name': 'nodejs',
      'process.runtime.description': 'Node.js',
      'telemetry.sdk.language': 'nodejs',
      'telemetry.sdk.name': 'opentelemetry',
      'telemetry.sdk.version': '2.0.1'
    }
  },
  instrumentationScope: {
    name: '@opentelemetry/instrumentation-express',
    version: '0.52.0',
    schemaUrl: undefined
  },
  traceId: '61e8960c349ca2a3a51289e050fd3b82',
  parentSpanContext: {
    traceId: '61e8960c349ca2a3a51289e050fd3b82',
    spanId: '631b666604f933bc',
    traceFlags: 1,
    traceState: undefined
  },
  traceState: undefined,
  name: 'request handler - /rolldice',
  id: 'd8fcc05ac4f60c99',
  kind: 0,
  timestamp: 1755719307779000,
  duration: 2801.5,
  attributes: {
    'http.route': '/rolldice',
    'express.name': '/rolldice',
    'express.type': 'request_handler'
  },
  status: { code: 0 },
  events: [],
  links: []
}
{
  resource: {
    attributes: {
      'host.arch': 'arm64',
      'host.id': '8FEBBC33-D6DA-57FC-8EF0-1A9C14B919F8',
      'process.pid': 12460,
      // ... some resource attributes elided ...
      'process.runtime.version': '22.17.1',
      'process.runtime.name': 'nodejs',
      'process.runtime.description': 'Node.js',
      'telemetry.sdk.language': 'nodejs',
      'telemetry.sdk.name': 'opentelemetry',
      'telemetry.sdk.version': '2.0.1'
    }
  },
  instrumentationScope: {
    name: '@opentelemetry/instrumentation-http',
    version: '0.203.0',
    schemaUrl: undefined
  },
  traceId: '61e8960c349ca2a3a51289e050fd3b82',
  parentSpanContext: undefined,
  traceState: undefined,
  name: 'GET /rolldice',
  id: '631b666604f933bc',
  kind: 1,
  timestamp: 1755719307777000,
  duration: 4705.75,
  attributes: {
    'http.url': 'https://:8080/rolldice',
    'http.host': 'localhost:8080',
    'net.host.name': 'localhost',
    'http.method': 'GET',
    'http.scheme': 'http',
    'http.target': '/rolldice',
    'http.user_agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:141.0) Gecko/20100101 Firefox/141.0',
    'http.flavor': '1.1',
    'net.transport': 'ip_tcp',
    'net.host.ip': '::ffff:127.0.0.1',
    'net.host.port': 8080,
    'net.peer.ip': '::ffff:127.0.0.1',
    'net.peer.port': 63067,
    'http.status_code': 200,
    'http.status_text': 'OK',
    'http.route': '/rolldice'
  },
  status: { code: 0 },
  events: [],
  links: []
}

生成的 span 跟踪了对 /rolldice 路由的请求的生命周期。

向该端点发送更多请求。片刻之后,您将在控制台输出中看到 metrics,例如以下内容:

查看示例输出
{
  descriptor: {
    name: 'http.server.duration',
    type: 'HISTOGRAM',
    description: 'Measures the duration of inbound HTTP requests.',
    unit: 'ms',
    valueType: 1,
    advice: {}
  },
  dataPointType: 0,
  dataPoints: [
    {
      attributes: {
        'http.scheme': 'http',
        'http.method': 'GET',
        'net.host.name': 'localhost',
        'http.flavor': '1.1',
        'http.status_code': 200,
        'net.host.port': 8080,
        'http.route': '/rolldice'
      },
      startTime: [ 1755719307, 782000000 ],
      endTime: [ 1755719482, 940000000 ],
      value: {
        min: 1.439792,
        max: 5.775,
        sum: 15.370167,
        buckets: {
          boundaries: [
               0,    5,    10,   25,
              50,   75,   100,  250,
             500,  750,  1000, 2500,
            5000, 7500, 10000
          ],
          counts: [
            0, 5, 1, 0, 0, 0,
            0, 0, 0, 0, 0, 0,
            0, 0, 0, 0
          ]
        },
        count: 6
      }
    },
    {
      attributes: {
        'http.scheme': 'http',
        'http.method': 'GET',
        'net.host.name': 'localhost',
        'http.flavor': '1.1',
        'http.status_code': 304,
        'net.host.port': 8080,
        'http.route': '/rolldice'
      },
      startTime: [ 1755719433, 609000000 ],
      endTime: [ 1755719482, 940000000 ],
      value: {
        min: 1.39575,
        max: 1.39575,
        sum: 1.39575,
        buckets: {
          boundaries: [
               0,    5,    10,   25,
              50,   75,   100,  250,
             500,  750,  1000, 2500,
            5000, 7500, 10000
          ],
          counts: [
            0, 1, 0, 0, 0, 0,
            0, 0, 0, 0, 0, 0,
            0, 0, 0, 0
          ]
        },
        count: 1
      }
    }
  ]
}
{
  descriptor: {
    name: 'nodejs.eventloop.utilization',
    type: 'OBSERVABLE_GAUGE',
    description: 'Event loop utilization',
    unit: '1',
    valueType: 1,
    advice: {}
  },
  dataPointType: 2,
  dataPoints: [
    {
      attributes: {},
      startTime: [ 1755719362, 939000000 ],
      endTime: [ 1755719482, 940000000 ],
      value: 0.00843049454565211
    }
  ]
}
{
  descriptor: {
    name: 'v8js.gc.duration',
    type: 'HISTOGRAM',
    description: 'Garbage collection duration by kind, one of major, minor, incremental or weakcb.',
    unit: 's',
    valueType: 1,
    advice: { explicitBucketBoundaries: [ 0.01, 0.1, 1, 10 ] }
  },
  dataPointType: 0,
  dataPoints: [
    {
      attributes: { 'v8js.gc.type': 'minor' },
      startTime: [ 1755719303, 5000000 ],
      endTime: [ 1755719482, 940000000 ],
      value: {
        min: 0.0005120840072631835,
        max: 0.0022552499771118163,
        sum: 0.006526499509811401,
        buckets: { boundaries: [ 0.01, 0.1, 1, 10 ], counts: [ 6, 0, 0, 0, 0 ] },
        count: 6
      }
    },
    {
      attributes: { 'v8js.gc.type': 'incremental' },
      startTime: [ 1755719310, 812000000 ],
      endTime: [ 1755719482, 940000000 ],
      value: {
        min: 0.0003403329849243164,
        max: 0.0012867081165313721,
        sum: 0.0016270411014556885,
        buckets: { boundaries: [ 0.01, 0.1, 1, 10 ], counts: [ 2, 0, 0, 0, 0 ] },
        count: 2
      }
    },
    {
      attributes: { 'v8js.gc.type': 'major' },
      startTime: [ 1755719310, 830000000 ],
      endTime: [ 1755719482, 940000000 ],
      value: {
        min: 0.0025888750553131105,
        max: 0.005744750022888183,
        sum: 0.008333625078201293,
        buckets: { boundaries: [ 0.01, 0.1, 1, 10 ], counts: [ 2, 0, 0, 0, 0 ] },
        count: 2
      }
    }
  ]
}

下一步

使用 手动 instrumentation 丰富您自动生成的 instrumentation,以获取自定义的可观测性数据。

您还需要配置一个合适的 exporter,以便将 您的遥测数据导出到一个或多个遥测后端。

如果您想探索更复杂的示例,请查看 OpenTelemetry Demo,其中包括基于 JavaScript 的 Payment Service 和基于 TypeScript 的 Frontend Service

故障排除

出现问题了吗?您可以启用诊断日志记录以验证 OpenTelemetry 是否已正确初始化。

/*instrumentation.ts*/
import { diag, DiagConsoleLogger, DiagLogLevel } from '@opentelemetry/api';

// For troubleshooting, set the log level to DiagLogLevel.DEBUG
diag.setLogger(new DiagConsoleLogger(), DiagLogLevel.INFO);

// const sdk = new NodeSDK({...
/*instrumentation.mjs*/
import { diag, DiagConsoleLogger, DiagLogLevel } from '@opentelemetry/api';

// For troubleshooting, set the log level to DiagLogLevel.DEBUG
diag.setLogger(new DiagConsoleLogger(), DiagLogLevel.INFO);

// const sdk = new NodeSDK({...