Fly.io中Docker运行Puppeteer的依赖缺失与启动失败问题排查

部署在Fly.io的Node.js应用，使用Puppeteer抓取多网站搜索结果，本地运行测试均正常，但部署后崩溃。核心报错为libnss3.so文件缺失，尝试过Puppeteer官方示例Dockerfile、zenika/alpine-chrome、node:slim等基础镜像，均出现找不到依赖或Chrome/Chromium可执行文件的问题，怀疑Fly.io未正确使用自定义Dockerfile。

附相关文件 #

1. 当前Dockerfile #

FROM node:18-slim

WORKDIR /app

COPY package*.json ./
RUN npm install

RUN apt-get update && apt-get install -y \
    chromium \
    libnss3 \
    libatk1.0-0 \
    libatk-bridge2.0-0 \
    libcups2 \
    libdrm2 \
    libxkbcommon0 \
    libxcomposite1 \
    libxdamage1 \
    libxrandr2 \
    libgbm1 \
    libasound2 \
    libpangocairo-1.0-0 \
    libxshmfence1

COPY . .

EXPOSE 3000

CMD ["npm", "start"]

2. fly.toml配置 #

app = "your-app-name"
primary_region = "iad"

[http_service]
  internal_port = 3000
  force_https = true
  auto_stop_machines = true
  auto_start_machines = true
  min_machines_running = 0
  processes = ["app"]

3. Puppeteer启动代码 #

const puppeteer = require('puppeteer');

async function scrape() {
  const browser = await puppeteer.launch({
    headless: 'new',
    executablePath: process.env.CHROMIUM_PATH || '/usr/bin/chromium',
    args: [
      '--no-sandbox',
      '--disable-setuid-sandbox',
      '--disable-dev-shm-usage',
      '--disable-gpu',
      '--no-zygote'
    ]
  });
  const page = await browser.newPage();
  // 抓取逻辑...
  await browser.close();
}

4. 完整报错日志 #

Error: Failed to launch the browser process!
/path/to/node_modules/puppeteer/.local-chromium/linux-XXXXXX/chrome-linux/chrome: error while loading shared libraries: libnss3.so: cannot open shared object file: No such file or directory

TROUBLESHOOTING: https://pptr.dev/troubleshooting

    at onClose (/app/node_modules/puppeteer/lib/cjs/puppeteer/node/BrowserRunner.js:293:20)
    at ChildProcess.<anonymous> (/app/node_modules/puppeteer/lib/cjs/puppeteer/node/BrowserRunner.js:285:24)
    at ChildProcess.emit (node:events:513:28)
    at ChildProcess._handle.onexit (node:child_process:293:12)

1. 修正Dockerfile，确保依赖完整且路径正确 #

使用node:18-bookworm（Debian 12基础镜像，依赖支持更完善），明确安装所有必要系统依赖，并通过环境变量指定Puppeteer使用系统级Chromium：

FROM node:18-bookworm

WORKDIR /app

# 安装完整的系统依赖与Chromium
RUN apt-get update && apt-get install -y --no-install-recommends \
    chromium \
    libnss3 \
    libatk1.0-0 \
    libatk-bridge2.0-0 \
    libcups2 \
    libdrm2 \
    libxkbcommon0 \
    libxcomposite1 \
    libxdamage1 \
    libxrandr2 \
    libgbm1 \
    libasound2 \
    libpangocairo-1.0-0 \
    libxshmfence1 \
    libpango-1.0-0 \
    libcairo2 \
    libatspi2.0-0 \
    libx11-xcb1 \
    libxcb-dri3-0 \
    && rm -rf /var/lib/apt/lists/*

# 设置环境变量，跳过Puppeteer自带Chromium下载，指定系统安装路径
ENV PUPPETEER_EXECUTABLE_PATH=/usr/bin/chromium
ENV PUPPETEER_SKIP_CHROMIUM_DOWNLOAD=true

COPY package*.json ./
RUN npm install

COPY . .

EXPOSE 3000

CMD ["npm", "start"]

2. 简化Puppeteer启动配置 #

无需手动指定executablePath，直接读取环境变量即可：

const puppeteer = require('puppeteer');

async function scrape() {
  const browser = await puppeteer.launch({
    headless: 'new',
    args: [
      '--no-sandbox',
      '--disable-setuid-sandbox',
      '--disable-dev-shm-usage',
      '--disable-gpu',
      '--no-zygote',
      '--disable-software-rasterizer'
    ]
  });
  const page = await browser.newPage();
  // 抓取逻辑...
  await browser.close();
}

3. 验证Fly.io部署流程 #

• 确保使用fly deploy命令部署，而非Fly.io自动构建器（若之前配置过构建器，需删除fly.toml中[build]相关配置）
• 查看部署日志：执行fly logs，确认Dockerfile中的apt安装步骤完整执行，无报错
• 部署后进入机器验证：执行fly ssh console，运行which chromium确认可执行文件路径，运行ldd /usr/bin/chromium | grep libnss3验证依赖存在

4. 重新部署测试 #

执行以下命令清除缓存并重新部署：

fly deploy --no-cache

内容的提问来源于stack exchange，提问作者bluenaxela