We're seeing a significant increase in 500 errors using the Dropbox API, and very little indication about what might be going wrong. It appears that our requests are valid, because we can simply reissue the exact same request and it will succeed, whether by using `fetch` in the browser or `curl` on the command line. We've had failures in `copy_v2` and `continue` and I think others, all following this same pattern. I've seen other support requests have included the x-dropbox-request-id, so here are two examples that produced 500 errors:
6d61d6bd21ab488289e05b0c593fdf13
e00b0281c9624f8a899af1ec2f33bde3
Both succeeded when retried a fraction of a second later.
We'd appreciate any insight.
Thanks!