App review automation issue?

We submitted our app for review but were told that our Login button was unresponsive. We went through a lot of trial and error across different devices, OSes, and even matched the device apple tested on. Retried submission after some changes and failed again--unresponsive login. So we opened a code-level ticket and told them about our issue with the login and how we thought it might be a mistake on apple's side. When we looked in our API logs we saw that apple's testing was indeed successful and they visited several screens. Only after we showed them the evidence in this ticket did our app get approved shortly thereafter.

Was this an issue with an automated system initially? Has anyone else had this problem? We just want to try to avoid it in the future. I don't like the thought of having to do review and also a code-level ticket.

Apple is very specific about those failures. They didn't say the login didn't work, they it was "unresponsive".

The most likely explanation was that Apple was testing in an unusual network configuration to see how the app would respond. It didn't.

To avoid this problem make sure you handle all edge cases and all kinds of failure. At a basic level with TCP networking, there is no guarantee of any response. So every TCP network operation should have some kind of timeout.

And don't take your final approval as confirmation that you were doing it correctly all along. Maybe someone at Apple looked at your app and decided it wasn't too bad and to just forget it for now. That might not always be the case. It's very common for people to get rejections out of the blue for existing apps that have already been approved.

It sounds like you've already given Apple a kill switch whenever they choose to push it at some point in the future. I recommend you take this opportunity to fix the problem rather than polling all the other people in this forum who are also struggling with rejections.

App review automation issue?
 
 
Q