“file uri scheme”

I suspect it was a FILE_NOTIFY_CLOSE. Are you sure it isn't an auth event? Also, are you using separate es_client_t's for your notify and auth events? The timing issue I outlined is theoretically possible in both cases, but MUCH easier to do with auth events. It would be insightful to know if APFS triggers that or related ES events when decompressing a file. Open is what typically triggers decompression; however, in this case, revisiond was actually deleting a file (by calling remove on its path), so it's pretty likely that they had closed it just prior to the remove call. I will submit unified logs that cover the period in question, but your point remains. I'll try and take a look, but I'm not necessarily optimistic. Frankly, these issues generally have a high degree of uncertainty about them, so my comment there was really about acknowledging that reality. Not that our ES hooks are directly causing the deadlock, but that by hooking certain OS events and quickly acting on them, we're exerci

I did file FB 22349771 (space intentionally added to avoid moderation flag).

it's absolutely wild that literally years after this issue was raised, it still hasn't been resolved. Even temporarily disabling SIP is an unacceptable security risk in order to reclaim orphaned files taking up tens or even hundreds of GB.

Just in case, the following post provides details about filing feedback for the Foundation Models framework: Provide actionable feedback for the Foundation Models framework and the on-device LLM. Best, —— Ziqiao Chen  Worldwide Developer Relations.

Just in case, the following post provides details about filing feedback for the Foundation Models framework: Provide actionable feedback for the Foundation Models framework and the on-device LLM. Best, —— Ziqiao Chen  Worldwide Developer Relations.

Given that you'd move the Core Data store to a group container, I'd assume that the store will be shared across different processes, such as your main app and its extension (a widget, for example). In that case, I'd like to start with pointing you the following threads: Core Data Light Migration Crash When Widget is Installed (Error 134100). CoreData error=134100 Failed to open the store. These threads made it clear that: After a user installed your new version app, depending on what extension you have, the app and extension can open and migrate the store simultaneously, and so you will need to implement a mechanism to make sure that the store is only migrated once. Consider having your main app in charge of the the synchronization, as discussed in Avoid synchronizing a store with multiple persistent containers. Now to your questions: Store relocation vs schema migration Is it advisable to perform store relocation and schema migration in a single step, or should these be separate releases? If combined, are th

Thanks for bringing this to the forums. [quote='820659021, idopte, /thread/820659, /profile/idopte'] returns the following error: Domain: CryptoTokenKit Code: -2 [/quote] That is TKErrorCodeCommunicationError, aka TKError.Code.communicationError in Swift. This is a very generic error that basically means that something went wrong with… well… the communication with the smart card. [quote='820659021, idopte, /thread/820659, /profile/idopte'] At this point, the system appears to be stuck in a non-recoverable state which affects all readers and cards [/quote] Yeah, that’s not good. This is clearly a bug and I appreciate you filing a bug report about it (FB22339746). I took a look at your bug and noticed that you haven’t attached a sysdiagnose to it. Or, more accurately, it says that there’s an associated sysdiagnose log but it hasn’t finished uploading. Please run Feedback Assistant on the iOS device that captured the sysdiagnose log and check on its upload status. Alternatively, export the sysdiagnose l

Hello, Additional information: the same subscription works correctly with StoreKit local testing in Xcode (StoreKit configuration file). The product is returned and the purchase flow behaves as expected. The issue only occurs in the App Store sandbox via TestFlight, where the same product ID returns an empty array. This suggests the implementation is correct and the issue may be related to App Store Connect or sandbox availability. Thank you.