I have an iOS app with ExtensionFoundation. It runs well on my local device, but when I upload on the AppStore it gets rejected with: Validation failed Invalid Info.plist value. The value of the EXExtensionPointIdentifier key, AsheKube.app.a-Shell.localWebServer, in the Info.plist of “a-Shell.app/Extensions/localWebServer.appex” is invalid. Please refer to the App Extension Programming Guide at https://developer.apple.com/library/content/documentation/General/Conceptual/ExtensibilityPG/Action.html#/apple_ref/doc/uid/TP40014214-CH13-SW1. (ID: ae8dd1dd-8caf-4a48-9651-7a225faed4eb) The Info.plist in my Extension is: <?xml version="1.0" encoding="UTF-8"?> <!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd"> <plist version="1.0"> <dict> <key>EXAppExtensionAttributes</key> <dict> <key>EXExtensionPointIdentifier</key> <string>com.example.example-extension</string> </dict> </dict> </plist> so the Info.plist that causes the issue has been automatically generated by Xcode. I can access it as well, and it says: { "BuildMachineOSBuild" => "25A354" "CFBundleDevelopmentRegion" => "en" "CFBundleDisplayName" => "localWebServerExtension" "CFBundleExecutable" => "localWebServer" "CFBundleIdentifier" => "AsheKube.app.a-Shell.localWebServerExtension" "CFBundleInfoDictionaryVersion" => "6.0" "CFBundleName" => "localWebServer" "CFBundlePackageType" => "XPC!" "CFBundleShortVersionString" => "1.0" "CFBundleSupportedPlatforms" => [ 0 => "iPhoneOS" ] "CFBundleVersion" => "1" "DTCompiler" => "com.apple.compilers.llvm.clang.1_0" "DTPlatformBuild" => "23A339" "DTPlatformName" => "iphoneos" "DTPlatformVersion" => "26.0" "DTSDKBuild" => "23A339" "DTSDKName" => "iphoneos26.0" "DTXcode" => "2601" "DTXcodeBuild" => "17A400" "EXAppExtensionAttributes" => { "EXExtensionPointIdentifier" => "AsheKube.app.a-Shell.localWebServer" } "MinimumOSVersion" => "26.0" "NSHumanReadableCopyright" => "Copyright © 2025 AsheKube. All rights reserved." "UIDeviceFamily" => [ 0 => 1 1 => 2 ] "UIRequiredDeviceCapabilities" => [ 0 => "arm64" ] } What should I do to be able to upload on the AppStore?

Hello! I am attempting to add Subscriptions to an App that Is already published on the App Store. I cannot get Xcode to actually sync what is in my App Store Connect. When adding the Storekit configuration file, I go through the automatic linking process and select the proper bundleID. The configuration file says 'Synced @ [CurrentTime]' however there are no subscriptions listed in there. I have attempted deleting the file several times, creating a new subscription group. With no success. Do I need to publish the subscriptions without the features first? Upon attempting to write the supporting code that will enable these features within the app, I cannot get Xcode to identify that I have these subscriptions. I have also tried pushing these to TestFlight, still with no success. Thank you.

I am developing a VoIP phone application(Our Phoneapp) using APNs VoIP push. I have a question regarding a behavior I discovered during testing of this application. When performing the following operations using an iPhoneSE3 with an sXGP-NW SIM inserted, 0xBAADCA11 occurs upon receiving an APNs VoIP PUSH. Do you have any information regarding this issue? 0xBAADCA11 occurs in operation 8. However, since there were no problems in operation 4 (the app works when Wi-Fi is off), I think there is no issue with the Our Phoneapp. [Configuration of system components] [VoIP Telephone] --Call to iPhone(Phoneapp)--> [Our VoIP PBX Server] -- VoIP PUSH request --> [Apple APNs Server] -- VoIP PUSH --> [Our Phoneapp (iPhoneSE3(with sXGP SIM)] [Operations] (The issue is reproducible 100% by following oparation) iPhoneSE3: Power on (iPhoneSE3 with sXGP SIM) iPhoneSE3: Wi-Fi off, connect to the internet via SIM. VoIP Telephone: Call to Our Phoneapp iPhoneSE3: Receives VoIP PUSH and Phoneapp launches. Successfully answers the call and communication is possible. (Receives VoIP push notification from APNs via sXGP SIM) iPhoneSE3: Wi-Fi is turned ON, connect to the internet via Wi-Fi. iPhoneSE3: Task kill Our Phoneapp. VoIP Telephone: Call to Our Phoneapp iPhoneSE3: iOS does not call the push notification delegate (didReceiveIncomingPushWithPayload). As a result our Phoneapp is unable to detect the incoming call, However, an ips log with 0xBAADCA11 is output. in other words, iOS received the VoIP PUSH, but Our Phoneapp dose not call CallKit, so Our Phoneapp was terminated by iOS.

Background Android phones supporting Wi-Fi Aware 4.0 should be able to connect with iPhones (iOS 26). For testing, we selected two Samsung S25 devices, which support Wi-Fi Aware 4.0. Issues we are facing Android as Publisher, iOS as Subscriber.iOS cannot discover the service. Log shows: Discovery: Dropping event, 02:14:60:76:a6:0f missing DCEA attribute. iOS as Publisher, Android as Subscriber,Android can discover the service.However, the PIN code is not displayed on iOS. From the packet capture, the publish packet does not contain the DCEA field. However, Android-to-Android devices can still pair normally, and the subsequent PASN packets include the DCEA field. It seems that the Wi-Fi Alliance only requires the DCEA to be present in the PASN packets. iOS cannot discover Android devices or complete pairing — is this caused by the DCEA field, or by other reasons?

Hi, We are developing an app using PacketTunnelProvider from Network Extension framework. It is packaged as a system extension. We are trying to implement an "always-on" functionality, but cannot manage to start the extension before user login, with or without on-demand enabled. However we see in other posts (1, 2) that a network extension packaged as sysex should automatically start before user login. Are we missing something? Is it a limitation of PacketTunnelProvider? Thanks

I just released an App update the didn't touch ANYTHING to do with Core Data (nothing changed in our Coredata code for at least 8 months). The update uses SDK for iOS 18 and Xcode 16.2 and the app now requires iOS 18 and was a minor bug patch and UI improvements for recent iOS changes. Since the update we are getting a steady trickle of users on iOS 18, some who allow the App to store data in iCloud (Cloudkit) and others who do not, all reporting that after the update to our recent release ALL their data is gone?! I had not seen this on ANY device until today when I asked a friend who uses the App if they had the issue and it turned out they did, so I hooked their device up to Xcode and ALL the data in the CoreData database was gone?! They are NOT using iCloud. There were no errors or exceptions on Xcode console but a below code returned NO records at all?! Chart is custom entity and is defined as: @interface Chart : NSManagedObject {} let moc = pc.viewContext let chartsFetch = NSFetchRequest<NSFetchRequestResult>(entityName:"Charts") // Fetch all Charts do { let fetchedCharts = try moc.fetch(chartsFetch) as! [Chart] for chart in fetchedCharts { .... } } A break point inside the do on fetchedCharts show there are NO objects returned. This is a serious issue and seems like an iOS 18 thing. I saw some people talking in here about NSFetchRequest issues with iOS 18. I need some guidance here from someone Apple engineer here who knows what the status of these NSFetchrequest bugs are and what possible workarounds are. Becasue this problem will grow for me as more users update to iOS 18.

Hi Folks, We are reading the USB device data from our app using libusb/iokit libraries. Before updating the MacOS to the 15.3 we never faced any issue but after updating OS to 15.3 Sequoia we started facing issue to access the USB device's information. We are not getting the device endpoints for the matching service and fails with below error- Error:Failed to create IOUSBHostObject. with reason: IOServiceOpen failed. Respective code snippet- service = IOServiceGetMatchingService(kIOMasterPortDefault, matchingDictionary); IOUSBHostInterface* interface = [[IOUSBHostInterface alloc] initWithIOService:service options:IOUSBHostObjectInitOptionsDeviceCapture queue:*queue error:&error interestHandler:nil]; We get the denial message during accessing the IOService error 23:17:30.691934-0800 kernel 41 duplicate reports for Sandbox: spotlightknowledged(1399) deny(1) mach-lookup com.apple.diagnosticd error 23:17:30.691945-0800 kernel System Policy: com.prograde.pgdrefreshpro.helpe(70515) deny(1) iokit-open-service IOUSBHostInterface Also when we checked the IOUSBHOST logs we can see pipes are stalled while running the RefreshPro app as below- 2025-02-05 22:06:31.838141-0800 0x25913e Error 0x0 0 0 kernel: (IOUSBHostFamily) AppleUSBIORequest: AppleUSBIORequest::complete: device 8 (SD PG05.5@08210000) endpoint 0x00: status 0xe0005000 (pipe stalled): 0 bytes transferred We need an assistance here to know what exactly could be the cause and how can we elevate the permissions to access the USB device on MacOS15.3. Do we need other entitlements? As we never faced such issue with our certificate and Identifier on any MacOS versions and with the current entitlements we have. Do we need to include any entitlement in the code? Thanks.

If I go to "System Settings" -> "General" -> "About", it says "MacBook Air" and below that "M2, 2022" How can I get these strings programatically? The following C code gets me #include <stdio.h> #include <sys/sysctl.h> int main() { static const unsigned namelen = 2; int name[namelen] = {CTL_HW, HW_PRODUCT}; char buffer[256]; size_t bufferSize = sizeof(buffer); if (0 != sysctl(name, namelen, buffer, &bufferSize, NULL, 0)) { perror("sysctl"); return 1; } printf("%s\n", buffer); } the string "Mac14,2", which is the hardware model identifier. But I want to get the user-friendly model name, e.g. "MacBook Air (13-inch, M2, 2022)". How can I do this?

I am looking for inputs to better understand MacOS entitlements. I ask this in context of OpenJDK project, which builds and ships the JDK. The build process makes uses of make tool and thus doesn't involving building through the XCode product. The JDK itself is a Java language platform providing applications a set of standard APIs. The implementation of these standard APIs internally involves calling platform specific native library functions. In this discussion, I would like to focus on the networking functions that the implementation uses. Almost all of these networking functions and syscalls that the internal implementation uses are BSD socket related. Imagine calls to socket(), connect(), getsockopt(), setsockopt(), getaddrinfo(), sendto(), listen(), accept() and several such. The JDK that's built through make is then packaged and made available for installation. The packaging itself varies, but for this discussion, I'll focus on the .tar.gz archived packaging. Within this archive there are several executables (for example: java, javac and others) and several libraries. My understanding, based on what I have read of MacOS entitlements is that, the entitlements are set on the executable and any libraries that would be loaded and used by that executable will be evaluated against the entitlements of the executable (please correct me if I misunderstand). Reading through the list of entitlements noted here https://developer.apple.com/documentation/bundleresources/entitlements, the relevant entitlements that an executable (like "java") which internally invokes BSD socket related syscalls and library functions, appear to be: com.apple.security.network.client - https://developer.apple.com/documentation/bundleresources/entitlements/com.apple.security.network.client com.apple.security.network.server - https://developer.apple.com/documentation/bundleresources/entitlements/com.apple.security.network.server com.apple.developer.networking.multicast - https://developer.apple.com/documentation/bundleresources/entitlements/com.apple.developer.networking.multicast Is my understanding correct that these are the relevant ones for MacOS? Are there any more entitlements that are of interest? Would it then mean that the executables (java for example) would have to enroll for these entitlements to be allowed to invoke those functions at runtime? Reading through https://developer.apple.com/documentation/bundleresources/entitlements, I believe that even when an executable is configured with these entitlements, when the application is running if that executable makes use of any operations for which it has an entitlement, the user is still prompted (through a UI notification) whether or not to allow the operation. Did I understand it right? The part that isn't clear from that documentation is, if the executable hasn't been configured with a relevant entitlement, what happens when the executable invokes on such operation. Will the user see a UI notification asking permission to allow the operation (just like if an entitlement was configured)? Or does that operation just fail in some behind the scenes way? Coming back to the networking specific entitlements, I found a couple of places in the MacOS documentation where it is claimed that the com.apple.developer.networking.multicast entitlement is only applicable on iOS. In fact, the entitlement definition page for it https://developer.apple.com/documentation/bundleresources/entitlements/com.apple.developer.networking.multicast says: "Your app must have this entitlement to send or receive IP multicast or broadcast on iOS. It also allows your app to browse and advertise arbitrary Bonjour service types." Yet, that same page, a few lines above, shows "macOS 10.0+". So, is com.apple.developer.networking.multicast entitlement necessary for an executable running on MacOS which deals with multicasting using BSD sockets? As a more general comment about the documentation, I see that the main entitlements page here https://developer.apple.com/documentation/bundleresources/entitlements categorizes some of these entitlements under specific categories, for example, notice how some entitlements are categorized under "App Clips". I think it would be useful if there was a category for "BSD sockets" and under that it would list all relevant entitlements that are applicable, even if it means repeating the entitlement names across different categories. I think that will make it easier to identify the relevant entitlements. Finally, more as a long term question, how does one watch or keep track of these required entitlements for these operations. What I mean is, is it expected that application developers keep visiting the macos documentation, like these pages, to know that a new entitlement is now required in a new macos (update) release? Or are there other ways to keep track of it? For example, if a newer macos requires a new entitlement, then when (an already built) executable is run on that version of macos, perhaps generate a notification or some kind of explicit error which makes it clear what entitlement is missing? I have read through https://developer.apple.com/documentation/bundleresources/diagnosing-issues-with-entitlements but that page focuses on identifying such issues when a executable is being built and doesn't explain the case where an executable has already been shipped with X entitlements and a new Y entitlement is now required to run on a newer version of macos.

The online documentation for fs_snapshot_create, which is on a website which apparently I'm not allowed to link to on this forum, mentions that some entitlement is necessary, but doesn't specify which one. Searching online I found someone mentioning com.apple.developer.vfs.snapshot, but when adding this to my entitlement file and building my Xcode project, I get the error Provisioning profile "Mac Team Provisioning Profile: com.example.myApp" doesn't include the com.apple.developer.vfs.snapshot entitlement. Searching some more online, I found someone mentioning that one has to request this entitlement from DTS. Is this true? I couldn't find any official documentation. I actually want to make a snapshot of a user-selected directory so that my app can sync it to another volume while avoiding that the user makes changes during the sync process that would make the copy inconsistent. Would fs_snapshot_create be faster than traversing the chosen directory and creating clones of each nested file with filecopy and the flag COPYFILE_CLONE? Although I have the impression that only fs_snapshot_create could make a truly consistent snapshot.

ICEcard app is "Emergency as a Service" platform. One of the key feature is to know about primary info, health info, or in case missing child , elderly using Face scan of registered user of app via another registered user of ICEcard app. App was working fine but last 2-3 week back got issue reported of app getting closed as soon Face scan option is selected. to simulate issue &gt; register &gt; open face scan icon at bottom home screen&gt; select any of option accident or health issue or information &gt;&gt; app closes immediately. Android app is working fine. link of app store. https://apps.apple.com/in/app/ice-card-app/id6736453602 android link for reference https://play.google.com/store/apps/details?id=com.rannlab.ice_card.ice_card&amp;pcampaignid=web_share

In my app the user can select a source folder to be synced with a destination folder. The sync can also happen in response to a change in the source folder detected with FSEventStreamCreate. If the user unzips an archive in the source folder and the sync process begins before the unzip operation has completed, the sync can fail because of a "Permission denied" error. I assume this is related to the posix permissions of the extracted folder being 420 during the unzip operation and (in my case) 511 afterwards. Is there a way to detect than an unzip operation is in progress and wait until it has completed? I thought that using NSFileCoordinator would solve this issue, but unfortunately it's not the case. Since an unzip operation can last any amount of time, it's not ideal to just delay a sync by a fixed number of seconds and let the user deal with any error if the unzip operation takes longer. let openPanel = NSOpenPanel() openPanel.canChooseDirectories = true if openPanel.runModal() == .cancel { return } let url = openPanel.urls[0].appendingPathComponent("extracted", isDirectory: false) var error: NSError? NSFileCoordinator(filePresenter: nil).coordinate(readingItemAt: url, error: &error) { url in do { print(try FileManager.default.attributesOfItem(atPath: url.path).sorted(by: { $0.key.rawValue < $1.key.rawValue }).map({ ($0.key.rawValue, $0.value) })) try FileManager.default.contentsOfDirectory(at: url, includingPropertiesForKeys: nil) } catch { print(error) } } if let error = error { print("file coordinator error:", error) }

On a MacBook Pro, 16GB of RAM, 500 GB SSD, OS Sequoia 15.7.1, M3 chip, I am running some python3 code in a conda environment that requires lots of RAM and sure enough, once physical memory is almost exhausted, swapfiles of about 1GB each start being created, which I can see in /System/Volumes/VM. This folder has about 470 GB of available space at the start of the process (I can see this through get info) however, once about 40 or so swapfiles are created, for a total of about 40GB of virtual memory occupied (and thus still plenty of available space in VM), zsh kills the python process responsible for the RAM usage (notably, it does not kill another python process using only about 100 MB of RAM). The message received is "zsh: killed" in the tmux pane where the logging of the process is printed. All the documentation I was able to consult says that macOS is designed to use up to all available storage on the startup disk (which is the one I am using since I have only one disk and the available space aforementioned reflects this) for swapping, when physical RAM is not enough. Then why is the process killed long before the swapping area is exhausted? In contrast, the same process on a Linux machine (basic python venv here) just keeps swapping, and never gets killed until swap area is exhausted. One last note, I do not have administrator rights on this device, so I could not run dmesg to retrieve more precise information, I can only check with df -h how the swap area increases little by little. My employer's IT team confirmed that they do not mess with memory usage on managed profiles, so macOS is just doing its thing. Thanks for any insight you can share on this issue, is it a known bug (perhaps with conda/python environments) or is it expected behaviour? Is there a way to keep the process from being killed?

I have an app that uses CKShare to allow users to share CloudKit data with other users. With the first build of the iOS 26, I'm seeing a few issues: I'm not able to add myself as a participant anymore when I have the link to a document. Some participants names no longer show up in the app. Looking at the release notes for iOS & iPadOS 26 Beta, there is a CloudKit section with two bullets: CloudKit sharing URLs do not launch third-party apps. (151778655) The request access APIs, such as CKShareRequestAccessOperation, are available in the SDK but are currently nonfunctional. (151878020) It sounds like the first issue is addressed by the first bullet, although the error message makes me wonder if I need to make changes to my iCloud account permissions or something in order to open it. It works fine in iOS 18.5. This is the error I get when I try to open a link to a shared document (I blocked out my email address, which is what was in quotes): As far as the second issue, I am really confused about what is going on. Some names still show up, while others do not. I can't find a pattern, and the missing users are not on the iOS 26 beta. The release notes mention CKShareRequestAccessOperation being nonfunctional, which is new in the beta and has some minor documentation, but I can't find information about how it's supposed to be used yet. In previous years there have been WWDC sessions about what's new in CloudKit, but I haven't found anything that talks about these changes to document sharing. Is there a guide or session somewhere that I'm missing? Does anyone know what's going on with these changes to CloudKit?

Hi all, We are migrating a SCSI HBA driver from KEXT to DriverKit (DEXT), with our DEXT inheriting from IOUserSCSIParallelInterfaceController. We've encountered a data corruption issue that is reliably reproducible under specific conditions and are hoping for some assistance from the community. Hardware and Driver Configuration: Controller: LSI 3108 DEXT Configuration: We are reporting our hardware limitations to the framework via the UserReportHBAConstraints function, with the following key settings: // UserReportHBAConstraints... addConstraint(kIOMaximumSegmentAddressableBitCountKey, 0x20); // 32-bit addConstraint(kIOMaximumSegmentCountWriteKey, 129); addConstraint(kIOMaximumByteCountWriteKey, 0x80000); // 512KB Observed Behavior: Direct I/O vs. Buffered I/O We've observed that the I/O behavior differs drastically depending on whether it goes through the system file cache: 1. Direct I/O (Bypassing System Cache) -> 100% Successful When we use fio with the direct=1 flag, our read/write and data verification tests pass perfectly for all file sizes, including 20GB+. 2. Buffered I/O (Using System Cache) -> 100% Failure at >128MB Whether we use the standard cp command or fio with the direct=1 option removed to simulate buffered I/O, we observe the exact same, clear failure threshold: Test Results: File sizes ≤ 128MB: Success. Data checksums match perfectly. File sizes ≥ 256MB: Failure. Checksums do not match, and the destination file is corrupted. Evidence of failure reproduced with fio (buffered_integrity_test.fio, with direct=1 removed): fio --size=128M buffered_integrity_test.fio -> Test Succeeded (err=0). fio --size=256M buffered_integrity_test.fio -> Test Failed (err=92), reporting the following error, which proves a data mismatch during the verification phase: verify: bad header ... at file ... offset 1048576, length 1048576 fio: ... error=Illegal byte sequence Our Analysis and Hypothesis The phenomenon of "Direct I/O succeeding while Buffered I/O fails" suggests the problem may be related to the cache synchronization mechanism at the end of the I/O process: Our UserProcessParallelTask_Impl function correctly handles READ and WRITE commands. When cp or fio (buffered) runs, the WRITE commands are successfully written to the LSI 3108 controller's onboard DRAM cache, and success is reported up the stack. At the end of the operation, to ensure data is flushed to disk, the macOS file system issues an fsync, which is ultimately translated into a SYNCHRONIZE CACHE SCSI command (Opcode 0x35 or 0x91) and sent to our UserProcessParallelTask_Impl. We hypothesize that our code may not be correctly identifying or handling this SYNCHRONIZE CACHE opcode. It might be reporting "success" up the stack without actually commanding the hardware to flush its cache to the physical disk. The OS receives this "success" status and assumes the operation is safely complete. In reality, however, the last batch of data remains only in the controller's volatile DRAM cache and is eventually lost. This results in an incomplete or incorrect file tail, and while the file size may be correct, the data checksum will inevitably fail. Summary Our DEXT driver performs correctly when handling Direct I/O but consistently fails with data corruption when handling Buffered I/O for files larger than 128MB. We can reliably reproduce this issue using fio with the direct=1 option removed. The root cause is very likely the improper handling of the SYNCHRONIZE CACHE command within our UserProcessParallelTask. P.S. This issue did not exist in the original KEXT version of the driver. We would appreciate any advice or guidance on this issue. Thank you.

We would be creating N NWListener objects and M NWConnection objects in our process' communication subsystem to create server sockets, accepted client sockets on server and client sockets on clients. Both NWConnection and NWListener rely on DispatchQueue to deliver state changes, incoming connections, send/recv completions etc. What DispatchQueues should I use and why? Global Concurrent Dispatch Queue (and which QoS?) for all NWConnection and NWListener One custom concurrent queue (which QoS?) for all NWConnection and NWListener? (Does that anyways get targetted to one of the global queues?) One custom concurrent queue per NWConnection and NWListener though all targetted to Global Concurrent Dispatch Queue (and which QoS?)? One custom concurrent queue per NWConnection and NWListener though all targetted to single target custom concurrent queue? For every option above, how am I impacted in terms of parallelism, concurrency, throughput &amp; latency and how is overall system impacted (with other processes also running)? Seperate questions (sorry for the digression): Are global concurrent queues specific to a process or shared across all processes on a device? Can I safely use setSpecific on global dispatch queues in our app?

I am encountering an issue with my application, BloxOneEndpoint.pkg, which includes two services: rc_service_infoblox – Runs as the root user. Controller Application – Runs as a normal user. Although a thread within rc_service_infoblox is running fine and performing its expected tasks, I notice that the service appears as "Not Responding" in Activity Monitor. Despite normal functionality, this status is concerning, as it may indicate some issue to customer. I would appreciate any insights into why this might be happening and how to resolve it. Is there a specific API or mechanism I should use to ensure the service remains in a "Running" state in Activity Monitor? Thank you for your guidance.

When I tap on one of the buttons in the ShieldAction extension I want to close the shield and open the parent app instead of the shielded app. Is there any way of doing this using the Screen Time API? class ShieldActionExtension: ShieldActionDelegate {      override func handle(action: ShieldAction, for application: ApplicationToken, completionHandler: @escaping (ShieldActionResponse) -> Void) {     // Handle the action as needed.           let store = ManagedSettingsStore()               switch action {     case .primaryButtonPressed:       //TODO - open parent app       completionHandler(.defer)     case .secondaryButtonPressed:       //remove shield       store.shield.applications?.remove(application)       completionHandler(.defer)         @unknown default:       fatalError()     }   }   }

I've been wondering what is the memory limit for network extensions. Specifically, I'm using the NEPacketTunnelProvider extension point.The various posts on this forum mention 5 MB and 6 MB for 32-bit and 64-bit respectively. However I find that (at least on iOS 10) the upper limit seems to be 15 MB. Is this the new memory limit for extensions?

I can develop a PacketTunnelProvider on Mac with xcode. I work with my self codesign. But when I sign it with Developer ID after read https://developer.apple.com/forums/thread/737894 , it still fail when I turn on the vpn .