"This Edge Function has Crashed" - intermittent errors

Support
1

We’ve been getting this intermittent error shown to users (ecommerce website) for a year now - it doesn’t happen often which is why we haven’t looked into it. We’re also not sure why it’s caused. The edge function logs show the following errors:
Oct 24, 12:31:57 PM: error Uncaught (in promise) resource closed
Oct 23, 01:21:58 PM: 01JAWN77 error Unable to render error page
Oct 23, 01:21:58 PM: 01JAWN77 error resource closed

Note that the first line " error Uncaught (in promise) resource closed" appears about 15-20 times but I removed the duplicates. I do see it every time the error “This edge function has crashed” occurs.

The “unable to render error page” is seen less often. There are also other less frequent errors like:

Oct 23, 07:25:06 PM: 01JAXA05 error BadResource: Bad resource IDOct 23, 07:25:06 PM: 01JAXA05 error at Object.pull (ext:deno_web/06_streams.js:946:38)Oct 23, 07:25:06 PM: 01JAXA05 error at Module.invokeCallbackFunction (ext:deno_webidl/00_webidl.js:981:16)Oct 23, 07:25:06 PM: 01JAXA05 error at ReadableByteStreamController.pullAlgorithm (ext:deno_web/06_streams.js:3563:14)Oct 23, 07:25:06 PM: 01JAXA05 error at readableByteStreamControllerCallPullIfNeeded (ext:deno_web/06_streams.js:1242:49)Oct 23, 07:25:06 PM: 01JAXA05 error at ReadableByteStreamController.[[[PullSteps]]] (ext:deno_web/06_streams.js:5990:5)Oct 23, 07:25:06 PM: 01JAXA05 error at readableStreamDefaultReaderRead (ext:deno_web/06_streams.js:2513:36)Oct 23, 07:25:06 PM: 01JAXA05 error at ext:deno_web/06_streams.js:2742:9Oct 23, 07:25:06 PM: 01JAXA05 error at new Promise ()Oct 23, 07:25:06 PM: 01JAXA05 error at ext:deno_web/06_streams.js:2741:14Oct 23, 07:25:06 PM: 01JAXA05 error at eventLoopTick (ext:core/01_core.js:168:7) {Oct 23, 07:25:06 PM: 01JAXA05 error name: "BadResource"Oct 23, 07:25:06 PM: 01JAXA05 error }Oct 23, 07:56:21 PM: 01JAXBSB error Unable to render error pageOct 23, 07:56:21 PM: 01JAXBSB error resource closed

I realise I may be reporting multiple issues here - the common theme is: I have not idea where the error is or what’s causing it.

One thing I noticed is that requests that may take longer than a few seconds may result in the most frequent error which is “error Uncaught (in promise) resource closed” but I might be wrong.

Any idea where I should start to look?

Thanks

2

Potentially related: Netlify Status - Increased errors and latencies across the network

3

This is likely related.
But I do get this issue at least once a day. Is it a memory allocation issue?
Is there a way for me to know if bandwidth or machine allocation is the issue and find a way to resolve this?
We’re seriously considering migrating off of Netlify to Firebase App hosting which we had a more positive experience with for years. easy of use and maintenance of Netlify are great pluses - but this shouldn’t come at the cost of reliability.

4

It would be hard to say without looking a specific request outside of the incident’s time. What do you see in the logs when this happens?

5

We see “error Uncaught (in promise) resource closed”. If the error doesn’t ring a bell for a known root cause let me know - we can dig in more to find the culprit. The generic error message doesn’t help but I believe I can replicate this when I try to reload certain pages which do server-side API calls.

6

Just looking at 1 day of logs, I see the following error more than 100 times throughout the day:
Oct 30, 02:54:15 PM: error Uncaught (in promise) resource closed

Any idea where I should be looking to get more details about the error?
The app is a QwikJS app - with the server side doing (not-very-frequent) calls to a an API to fetch data.

7

Do you recommend that we move away from edge functions to better understand the issue? it seems like other face this with Deno (other issues, different platforms and hosting solutions.)

8

Can you share the site name OR URL so we can check?