[▲ Vercel Community](/) · [Categories](/categories) · [Latest](/latest) · [Top](/top) · [Live](/live) [Help](/c/help/9) # Vercel function crashes on any network request 129 views · 10 likes · 7 posts Sousandrei (@sousandrei) · 2025-04-14 · ♥ 1 <!-- Questions that get answered the fastest are the ones with relevant info included in the original post. Be sure to include all detail needed to let others see and understand the problem! --> Recently, my react-router project started failing on a POST request. It's weird since it does not do much, just a FETCH request. I then narrowed it down to, whenever FETCH gets called, something crashes hardcore and I can't even catch the error, no idea why. It works locally perfectly, but when deployed, errors The logs on the platform say nothing I've managed to reproduce the issue in a simple repo, basically created using Vercel's templates and added one commit https://github.com/sousandrei/react-router-bug/blob/main/app/routes/test.tsx The expected output is of course a `200` and both before and after console messages being printed, or in a bad case, the `ERROR` from the catch. We get the before but nothing after Using httpie ``` http --verify=no POST https://react-router-bug.vercel.app/test ``` then we get  PS: tried using axios as well, same issue CogniStream (@cognistream-admin) · 2025-04-14 · ♥ 3 I've been trying to debug this error too today. * Only happens for POST requests, not GET. * Old deployment still works. * New deployment made from the same commit fails. I deployed a patched version of `react-router` that disables `sanitizeError` and got this traceback: ``` DOMException [AbortError]: This operation was aborted at new DOMException (node:internal/per_context/domexception:53:5) at AbortController.abort (node:internal/abort_controller:465:18) at IncomingMessage.<anonymous> (/opt/rust/nodejs.js:2:1816) at Object.onceWrapper (node:events:632:28) at IncomingMessage.emit (node:events:518:28) at emitCloseNT (node:internal/streams/destroy:148:10) at process.processTicksAndRejections (node:internal/process/task_queues:89:21) ``` Also, the request does successfully complete, and by adding logging to the promise in the background I have sometimes got it to log the result of the request. It appears that it's just the response stream that is getting aborted. Jacob Paris (@jacobparis) · 2025-04-14 · ♥ 2 I can reproduce this, escalating internally Jacob Paris (@jacobparis) · 2025-04-14 This should be fixed now, try redeploying and let me know if it works for you Sousandrei (@sousandrei) · 2025-04-14 · ♥ 1 @jacobparis it does work now! Both in the test project and in the real one Thank you so much! Can you share what happened, just out of sheer curiosity? If not it's ok :smiley: CogniStream (@cognistream-admin) · 2025-04-14 · ♥ 1 Can also confirm it's now working. Would also love to hear what happened :slight_smile: Jacob Paris (@jacobparis) · 2025-04-14 · ♥ 2 Thanks for confirming! There was a new version of the runtime that made a few changes, which is why it only broke on new deploys but not old ones. I don't know too much beyond that yet
