I am not sure what you mean by 10.3. My current iOS and iPadOS versions are 13.3.1. I have gone through the entire set of iOS updates without the issue getting resolved (I usually update right away).
Oftentimes the issue seems to be resolved only to reappear a few weeks later. Additionally, the crashes can happen while the app is “idling”. My hunch is it is a buffer/cache of some sort that is getting filled up, Roon is not clearing out, and at some point iOS decides that it’s too full and kills the host app.
I wonder how much of the code used in the iOS app is shared with the iPadOS app (given that for one the interface is fairly different). I am wondering if there’s some issue in the iOS branch that is causing runaway buffers and forces iOS to kill the app.
Sorry, I meant 13.3 and have edited to clarify. I know that 13 in general has a much more aggressive approach to throttling/killing background apps, and wonder if that’s what’s at play here. It doesn’t quite fit, because the app is killed while in use, not backgrounded, but it’s easy to imagine that a similar mechanism is doing it.
Pretty sure iOS will kill an app it deems is using too much memory or cache space, no question. What is odd here is that I run Roon on both iOS 13.3.1 (official!!! ) and iPadOS 13.3.1 on an iPhone XS Max and iPad Pro both with 512GB of storage. So hardware is pretty much identical, but the iPadOS version is stable while the iOS version is a disaster.
So I am wondering whether it has something to do with the differences in the displays and the iOS branch of the code. It stinks as poor cache management on the iOS case. It reeks.
We discussed this issue once more during our meeting today. Our team continues to investigate this and we will be sure to update you as soon as we have more information available.
Unfortunately, this has been a bit of a tricky report to track down — We have a variety of iOS devices in our QA labs that we have been working to reproduce this issue with and it’s not something we are able to reproduce. A majority of our customers using these devices also remain unaffected by this issue.
It’s likely that there is something going on at an OS level that is causing this behavior to occur (we’ve seen reports come and go with various iOS updates), but again not everyone on the same version is impacted by this.
This remains a high priority issue for us that we are continuing to investigate and we will be sure to keep everyone updated as soon as we have any further information. In the meantime, if you can think of anything that makes your iOS device or your general setup different that might contribute to the reason you’re seeing this behavior more than any others please let us know and we’ll be sure to pass it along to the team.
I find that at times something clears out (possibly a cache) and Roon becomes stable for a while. Then it reaches a state (over days) where it starts becoming unstable and crashes all the time. It is not something that is contant and the same across the board, something is happening that makes it unstable at some point. Killing the app does not clear the instability, nor does clearing image cache or restarting the phone in my experience. Just more data points in case this didn’t come across.
I am sure you guys are greatly capable. If I were looking at this I would try and see what OS parameters (eg tmp cache or similar) persist across sessions in xcode. It’s got to be something that makes it unstable across sessions. It is quite possible that xcode is clearing out all caches once the simulation ends which might not be the case in the iPhone.
Just a thought:
I have been a Qobuz subscriber for four years. And during this time, the Qobuz app has become increasingly, and to a great extent, hungrier and in need of resources. You only need to look into Play Store or App Store:
But dozens of people, owners of older devices, who complain of constant crashes in the Qobuz app.
With my older smartphone, an LG V10, I can only manage just as well for the garden. As soon as I download something to a memory card, change playlists or do something that stresses the processor and consumes RAM, the Qobuz app crashes.
And apparently it’s not just me.
Okay …
why should it be different with Roon?
Regarding Qobuz, I bought another device with a stronger processor and more RAM …
I haven’t had any problems since then.
But maybe it is due to my network?
I do not Know.
In any case, I have none of the problems mentioned here - everything works like a charm without any difficulties. Turn on and play music.
I am not an expert. Just a music lover.
I run both an iPhone XS Max and an iPad Pro, identical amount of storage (512GB). Both running the same releases of software (iOS and iPadOS 13.3.1). And yet iPhone crashes whereas iPad does not.
My iPhone is stable for a while, then becomes very unstable. Then all of a sudden after a few days it will become stable again. Very strange and why I have the hunch it must be something related to cache management across sessions. It is also clear it is the OS killing it because otherwise the Roon app logs would tell you something (my logs were/are activated and Roon is getting the info from my phone). It also seems clear this is an issue with the iOS/phone part of the code (ie the GUI code used to render on iPhones).
I’ll also add that my understanding is xcode simulator is not great at keeping state between sessions or troubelshooting what happens with apps in the background, so this might be making it harder for the Roon team to debug. But a bug it sure is.
I just started having the same issue. I see that the earliest post here is from January. Everything has been working fine for me up until today. It’s crashing like crazy. It’s not really usable.
I’ve had a nice and stable two-three months and this issue began re-occurring for me again yesterday. Same as before, Roon on iPhone will crash after about thirty seconds. Deleting and reinstalling has not helped, and it also occurs in airplane mode as I saw suggested as a troubleshooting step in the split out thread linked above.
Is there any better understanding of this issue that the team can comment on?