Cannot download zip file into /tmp

Help
1

Current versus Expected behavior

I am considering deploying an app that scrapes a certain page.
I am trying to download Chromium to the /tmp folder using puppeteer-core.
However, when I attempt to download the zip file, the process stops without any errors.
I would like to ask about this issue.

Code, configuration, and steps that reproduce this issue

https://github.com/chihiro-yabuta/opas/blob/main/pages/api/scrape.ts#L19
The process after the pipe in this promise is stopping.

As additional information, this function is being executed asynchronously because it takes a long time.

Project information

Deployment URL: https://vercel.com/chihiro-yabutas-projects/opas/FsmSNi1ENmowwoCygN8LcFw3ZZyd
Environment (local, preview, production): production
Project Framework: nextjs
Build Settings:
  Framework Preset: nextjs
  Build Command (if not default):
  Output Directory (if not default):
  Install Command (if not default):
Node/Runtime Version:
Package Manager:
Relevant Packages: https, fs
2

Hi, @chihiro-yabuta! Welcome to the Vercel Community. :smile:

Vercel uses a serverless architecture, which has certain limitations . The /tmp directory in serverless environments is typically limited in size and not persistent between function invocations.

Installing Chromium or Puppeteer in a serverless environment is generally not recommended or supported. The installation process requires more resources and time than typically available in a serverless function execution.

It’s also worth pointing out from our our Terms of Service:

You may not use the Services or Vercel’s infrastructure for proxying, scraping, to create virtual private networks, or to create virtual private servers.

3

thank you. I’ll consider another way.