Just answering since you’ve gotten no bites yet.

Due to the fact that there is a default error message of “failed” on the client side, my guess is that that reason is always pretty generic. We’re chunking up to an S3 bucket so the only possibility for every day failure is a network error. We won’t be running out of space! Other possibilities like “account past due,” “missing API keys,” “bucket doesn’t exist,” etc are all things that you likely have ways of knowing immediately without having to wait for an upload to fail. Otherwise validations for things like file type and size are up to your own code before initiating the upload, so of course you have those errors.

This is just a hypothesis, though, as I’ve never looked into this myself—and maybe I’m missing something obvious? If you’re curious and are able to simulate an error, I would just hop into your deps directory and throw a dbg on that line and see if it says anything useful.