如何让错误栈追踪自定义代码调用位置而非AWS SDK内部抛出点？

代码示例 #

import { s3 } from 'aws-sdk'

async function usecase() {
  await myLogic1()
}

async function myLogic1() {
  //... my logic
  await s3.putObject()
}

错误栈信息 #

at Request.extractError (/Users/justinkao/working/AUTO-ops/mission-sticker/mission-sticker-pipeline/node_modules/aws-sdk/lib/services/s3.js:711:35)
    at Request.callListeners (/Users/justinkao/working/AUTO-ops/mission-sticker/mission-sticker-pipeline/node_modules/aws-sdk/lib/sequential_executor.js:106:20)
    at Request.emit (/Users/justinkao/working/AUTO-ops/mission-sticker/mission-sticker-pipeline/node_modules/aws-sdk/lib/sequential_executor.js:78:10)
    at Request.emit (/Users/justinkao/working/AUTO-ops/mission-sticker/mission-sticker-pipeline/node_modules/aws-sdk/lib/request.js:686:14)
    at Request.transition (/Users/justinkao/working/AUTO-ops/mission-sticker/mission-sticker-pipeline/node_modules/aws-sdk/lib/request.js:22:10)
    at AcceptorStateMachine.runTo (/Users/justinkao/working/AUTO-ops/mission-sticker/mission-sticker-pipeline/node_modules/aws-sdk/lib/state_machine.js:14:12)
    at /Users/justinkao/working/AUTO-ops/mission-sticker/mission-sticker-pipeline/node_modules/aws-sdk/lib/state_machine.js:26:10)
    at Request.<anonymous> (/Users/justinkao/working/AUTO-ops/mission-sticker/mission-sticker-pipeline/node_modules/aws-sdk/lib/request.js:38:9)
    at Request.<anonymous> (/Users/justinkao/working/AUTO-ops/mission-sticker/mission-sticker-pipeline/node_modules/aws-sdk/lib/request.js:688:12)
    at Request.callListeners (/Users/justinkao/working/AUTO-ops/mission-sticker/mission-sticker-pipeline/node_modules/aws-sdk/lib/sequential_executor.js:116:18)
    at Request.emit (/Users/justinkao/working/AUTO-ops/mission-sticker/mission-sticker-pipeline/node_modules/aws-sdk/lib/sequential_executor.js:78:10)
    at Request.emit (/Users/justinkao/working/AUTO-ops/mission-sticker/mission-sticker-pipeline/node_modules/aws-sdk/lib/request.js:686:14)

问题描述 #

调用s3.putObject抛出错误时，错误栈仅显示AWS SDK内部的函数调用，无法定位到自定义函数myLogic1()或usecase()的调用位置，如何让错误栈包含自定义代码的调用信息？

解决方案 #

方法1：手动捕获错误并保留调用栈 #

AWS SDK v2的方法默认使用回调模式，直接await s3.putObject()无法正确触发Promise链式调用，需先调用.promise()转为Promise，再通过try-catch捕获错误，创建包含当前调用栈的新错误并关联原始错误：

async function myLogic1() {
  //... my logic
  try {
    // 需传入合法参数，此处仅示例结构
    await s3.putObject({ Bucket: 'your-bucket', Key: 'your-key', Body: 'content' }).promise();
  } catch (err) {
    const customError = new Error(`执行S3上传失败: ${err.message}`);
    // 将原始错误作为cause保留，确保完整错误信息不丢失
    customError.cause = err;
    throw customError;
  }
}

抛出的自定义错误栈会包含myLogic1()和usecase()的调用位置，同时通过cause字段保留SDK原始错误细节。

方法2：使用Error.captureStackTrace补充栈信息 #

如果需要更精细的栈控制，可以手动捕获当前调用栈并附加到原始错误上：

async function myLogic1() {
  //... my logic
  const stackHolder = {};
  Error.captureStackTrace(stackHolder, myLogic1);
  
  try {
    await s3.putObject(params).promise();
  } catch (err) {
    // 将自定义栈拼接到原始错误栈末尾
    err.stack += '\n' + stackHolder.stack;
    throw err;
  }
}

方法3：升级到AWS SDK v3 #

AWS SDK v3采用模块化设计，原生支持Promise，错误栈会自动保留上层自定义函数的调用上下文。示例代码：

import { S3Client, PutObjectCommand } from "@aws-sdk/client-s3";

const s3Client = new S3Client({ region: 'your-region' });

async function myLogic1() {
  //... my logic
  const command = new PutObjectCommand({
    Bucket: 'your-bucket',
    Key: 'your-key',
    Body: 'content'
  });
  await s3Client.send(command);
}

此时抛出的错误栈会直接包含myLogic1()和usecase()的调用位置，无需额外处理。

内容的提问来源于stack exchange，提问作者高欣平