How frequently does the bug occur?

Sometimes

Description

A few days ago, we added a new object (model) which, according to the documentation, is a non-breaking change.

However, we have noticed a few users (~10% so far) triggering the sync manager's error handler with a clientUserError.
This can become an annoying behavior as we follow up with a logout and users have to re-login.

We also believe there is another related bug where the clientUserError isn't triggered instantaneously. I'll explain the flow :

1. User opens the app
2. App does a Realm.asyncOpen(configuration: config) (with config = partition-based and using recoverOrDiscardUnsyncedChanges) so that we are sure to be 100% sync every time. User is stuck on a loading screen meanwhile.
3. Upon success, user can finally continue with the app, and we set up one Realm listener on a specific object.
4. Client User Error is now triggered <- Why only now and not at step 2?

Unfortunately, we can't reproduce the bug on our side (+ it is too complex to set up the same conditions as our users who are crashing).

By the way, is there a way to manually trigger a clientUserError? How does the Realm team simulate this error in their unit test?

Stacktrace & log output

No response

Can you reproduce the bug?

No

Reproduction Steps

No response

Version

10.33.0

What Atlas Services are you using?

Both Atlas Device Sync and Atlas App Services

Are you using encryption?

Yes

Platform OS and version(s)

iOS 15.x and 16.x

Build environment

Xcode version: 14.2
Dependency manager and version: SPM