Dear Apple Developer Support Team, I am writing regarding critical issues we are facing with Safari web push notifications in our application iLiveMyLife.io, which is severely impacting our ability to maintain reliable communication with our users. Issue Description: We are experiencing persistent problems with Safari push notification tokens expiring or becoming invalid without any notification to our server. This creates several critical issues: Users stop receiving notifications without any indication of failure Our notification delivery system has no way to detect token expiration The expiration appears to happen frequently (seemingly almost daily in some cases) There is no reliable mechanism to re-establish push communication without users manually revisiting the app Technical Impact: Our messaging functionality becomes completely unreliable We must resort to email or SMS as fallback mechanisms, which is not feasible for a real-time communication platform This makes building any reliable messaging application on Safari practically impossible The Broader Context: What makes this situation particularly challenging is that all potential alternative browser APIs that could help address this issue appear to be deliberately disabled or restricted in Safari: Background Service Workers don't function in the background on iOS Safari Background Sync API is not supported WebSockets cannot operate when the app is closed There's no way to programmatically check the validity of push tokens The combination of these limitations creates a situation where developers have no viable technical path to build reliable notification systems for PWAs on Safari. This appears to be a systematic restriction rather than individual API limitations. Requested Information: Is there a recommended approach to detect Safari push token expiration? Are there alternative notification mechanisms for PWA applications on Safari that offer more reliability? Is there documentation on the lifecycle of Safari push tokens that could help us implement proper handling? Are there plans to improve the Web Push API implementation in Safari to address these reliability issues? Could you clarify if these limitations are intentional design decisions or technical constraints that might be addressed in future updates? Business Impact: This issue fundamentally undermines our platform's core functionality. For a collaborative tool, reliable notifications are essential - users cannot collaborate effectively if they miss updates because their push tokens silently expired. The current state creates confusion among our users, who don't understand why they suddenly stop receiving notifications. Any guidance or assistance you could provide would be greatly appreciated. We're committed to providing an excellent experience on Safari, but the current push notification limitations make this extremely challenging. Thank you for your time and consideration. Best regards, Ilya

We are currently developing a WebAPI service that uses APNs (HTTP/2) to send push notifications to iOS devices. (Using PushSharp's HTTP/2 support) The WebAPI service is running on IIS using .net framework 4.8 and c#. The connection to APNs is always maintained, and the connection is checked every 30 minutes using a dummy token Ping. KeepAlivePeriod = 30 minutes and KeepAliveRetryPeriod = 10 seconds have also been set. However, the following issues are occurring. Although the Ping sent immediately before was successful, a TimeoutException occurs in the notification request sent a few minutes later. There is no explicit disconnection notification from APNs, and the connection appears to be silently disconnected. Once a TimeoutException occurs, it occurs frequently afterwards. Below is an excerpt from the log. Apple Notification Failed for some unknown reason 1-1: One or more errors occurred. Apple Notification Failed for some unknown reason 1-2:System.TimeoutException: The operation timed out. In light of this issue, I would like to be advised on the following two points. Are there any official specifications regarding the lifecycle and expiration date of APNs HTTP/2 connections? Even if pings are sent periodically, is there a timeout or other setting that disables the connection on the APNs side? What are the conditions that would cause APNs to silently terminate a connection? For example, could this be due to inactivity, TLS restrictions, network maintenance, etc.? If you have any official documentation or technical guidelines to improve the reliability of this system, we would appreciate it if you could share them with us. Thank you in advance.

Back story: I'm developing an app that communicates with a personal medical device. We use critical alerts when we have hardware issues that could result in harm to the patient. The audio file is a 30 second file to make sure the patient is aware. If the app is open when they occur, we pop up a modal message in the app. When the user dismisses the notice, we call UNNotificationCenter::removeDeliveredNotifications(withIdentifiers:) to remove the critical alert and also to stop the audio file that is playing. This normally works fine. However we discovered that if the patient leaves critical alert enabled but disables notifications for our app, that we can still post the critical alert and it goes off. However when the user dismisses the message, the removeDeliveredNotifications call does not work. I did some debugging and if call getDeliveredNotifications with this permission combination, it return 0 (normally it would return 1). Does anyone know of another way to remove the critical alert in this situation? (or should I be submitting this as a bug?)

Hello, We are currently facing a persistent issue where Apple Push Notifications (APNs) are no longer being delivered to devices. This is not an intermittent problem — push notifications have completely stopped working. We are using PushSharp.Apple (Version 4.0.10.0) to send the notifications. Current Behavior (Failing): Every notification request appears to be processed successfully from the server side. However, the APNs client receives a 0-byte response, and the server closes the connection. Devices do not receive the notifications at all. Logs from Current (Failing) State: APNS-Client[1]: Connected (Batch ID=1) APNS-Client[1]: Sent Batch, waiting for possible response... APNS-Client[1]: Received 0 bytes response... APNS-Client[1]: Server Closed Connection... APNS-Client[1]: Disconnected (Batch ID=1) For Reference – Successful Log From Earlier (When Notifications Worked): APNS-Client[1]: Connected (Batch ID=1) APNS-Client[1]: Sent Batch, waiting for possible response... APNS-Client[1]: Received -1 bytes response... Push Notification Successfully Sent to Device Any help, suggestions, or experience with similar issues would be greatly appreciated. Thanks!!!

We are reaching out to report a critical issue we are facing with Apple Push Notification Service (APNs) for our application. Since yesterday, push notifications have stopped working, and we are consistently encountering connection reset errors when attempting to connect to Apple’s push notification servers. The issue specifically involves the following endpoints: gateway.push.apple.com – 17.188.170.74 feedback.push.apple.com – 17.188.178.87 These connections were functioning properly until yesterday. No changes have been made on our end, and we have verified our server configurations. We kindly request your assistance in investigating this matter. Please confirm whether there are any known issues or restrictions related to the above IPs or endpoints that could be affecting our service. This issue is impacting our production environment and affecting user experience, so your prompt support would be greatly appreciated.

requestUrl:https://api.storekit-sandbox.itunes.apple.com/inApps/v1/notifications/test method:Post responseCode:200 testNotificationToken:06beba6c-7587-44f6-a4b8-f12b11db6061_1745832671126 Request environment: ①APP from: TestFlight ②Apple acount : Sandbox account ③Apple originalTransactionId: 2000000907966801 App Store sandbox server notification configuration is fine(https://*.com/v2/apple_refund_notify), I have tried many times, but have not received any notification from Apple. Is this API not working?

Version: iOS 18.1 and later (works as expected on iOS 18.0 and earlier) Area: SpringBoard / Notification Center / App Icon Rendering Description: When changing the app's alternate icon using UIApplication.setAlternateIconName(_:completionHandler:), the icon is updated correctly on the Home Screen and App Switcher. However, in Notification Center, the old app icon is still shown for notifications, even after the change has completed. This issue only occurs on iOS 18.1 and later. In iOS 18.0 and earlier, Notification Center correctly reflects the updated icon. - Steps to reproduce: Create an iOS app with alternate app icons configured in the Info.plist. Use UIApplication.shared.setAlternateIconName("IconName") to change the icon at runtime. Send a notification. Pull down Notification Center and observe the icon shown beside the notification. - Expected Behavior: Notification Center should reflect the updated (alternate) app icon immediately after the change. - Actual Behavior: Notification Center continues to display the old (primary) app icon. The new icon appears correctly on the Home Screen and App Switcher. Restarting the device does cause Notification Center to update and reflect the correct icon, which suggests a cache or refresh issue in SpringBoard or Notification Center. - Notes: Issue introduced in iOS 18.1; not present in 18.0. Reproduces on both physical devices and simulators. Occurs with both scheduled local notifications and remote notifications. Restarting the device updates the Notification Center icon, but this is not a viable user-facing workaround.

When changing the app's alternate icon using UIApplication.setAlternateIconName(_:completionHandler:), the icon is updated correctly on the Home Screen and App Switcher. However, in Notification Center, the old app icon is still shown for notifications, even after the change has completed. Rebooting or change the device's language causes the correct icon to appear. This issue only occurs on iOS 18.1 and later. In iOS 18.0 and earlier, Notification Center correctly reflects the updated icon. Could you provide insights into how iOS caches notification icons and how we can force a refresh for all users?

I’m using Appnotic from my server to send notifications for an emergency service, where it is critical that notifications are delivered immediately. My payload looks like this: "aps": { "alert": "Test alert", "sound": { "critical": 1, "name": "sound.wav", "volume": 0.5 }, "content-available": 1, "category": "alert" }, "topic": "com.fireservicerota.FSR-Primary-Alerting", "custom_payload": { "id": "11", "type": "alert", "incident_id": 23434, "incident_response_id": 2652343, "expiration_time": "2024-06-06T16:59:05+01:00" } } I already have the critical alert entitlement and background processing enabled. Everything seems fine when debugging, but I’m experiencing issues: • Some notifications never arrive • Around 60% of notifications arrive with noticeable delay Since this is an emergency app, delivery speed is crucial. What could be causing this inconsistency?

I am checking if the user taps on the firebase push notification and get the payload. override func userNotificationCenter(_ center: UNUserNotificationCenter, didReceive response: UNNotificationResponse, withCompletionHandler completionHandler: @escaping () -> Void) { let userInfo = response.notification.request.content.userInfo os_log("notification tapped %{public}@", log: OSLog.push, type: .info, userInfo) handleNotificationPayload(userInfo as! [String: AnyObject]) setFlutterLinkClickedVariable() } My use case is in app terminated state when push notification is tapped, get the link from payload and navigate to corresponding screen based on the link. This is working when there is only one push notification. When there are multiple push notifications with different links in the payload, only the first notification I tap works. Rest of the notifications just launches the app and does not navigate because the link is not set. I am getting the link from the payload and invoking flutter code which sets the link in the user defaults (shared preferences) and when the app launches in the home screen it checks for this variable and navigates accordingly. func handleNotificationPayload(_ payload: [String: AnyObject]) { if let link = payload["link"] as? String { setFlutterLinkVariable(link) } } override func application(_ application: UIApplication, didReceiveRemoteNotification userInfo: [AnyHashable : Any], fetchCompletionHandler completionHandler: @escaping (UIBackgroundFetchResult) -> Void) { os_log("app did receive remote notification %{public}@", log: OSLog.push, type: .info, userInfo) handleNotificationPayload(userInfo as! [String : AnyObject]) completionHandler(.newData) } Currently when there is only one push notification it works because the link is set from the above method. The click delegate is not calling. I did set the delegate in application(:didFinishLaunchingWithOptions). UNUserNotificationCenter.current().delegate = self application.registerForRemoteNotifications() How to solve this issue? Thanks.

I’m building a companion app that connects to a custom hardware hub (IoT device) used for home safety monitoring. The hub is installed in homes and is responsible for triggering critical alerts like fire alarms, motion detection, door sensor activity, and baby monitor events. Current Architecture: The hub initially connects to the app via Bluetooth (BLE) for provisioning (to get Wi-Fi credentials). Once provisioned, the hub switches to Wi-Fi and communicates with the app via a WebSocket connection to stream real-time event updates. What I’m Trying to Achieve: My goal is to maintain background communication with the hub even when the app is not actively in use, in order to: Receive real-time updates from the hub while the device is locked or the app is in background. Trigger local notifications immediately when critical sensor events (e.g., fire, motion) occur. Ensure persistence across backgrounding, app swipes (force close), and device reboots, if possible. What I'm Observing: On iOS, WebSocket connection is suspended or dropped shortly after the app goes to the background or is locked. Even though the I've scheduled periodic fetches, notifications are delayed until the app is reopened, at which point all missed WebSocket messages arrive at once. If the app is force-closed or after reboot, no reconnection or notification happens at all. Key Questions I Have: Since the hub is initially provisioned via BLE, and could potentially send BLE flags or triggers for key events, can I use bluetooth-central mode to keep the app active or wake it up on BLE activity? Once the hub switches to Wi-Fi and uses WebSocket, is it possible to combine BLE triggers to wake the app and then reconnect to the WebSocket to fetch the full event payload? Is there a legitimate and App Store-compliant way to maintain a connection or background task with: BLE accessory triggers followed by Real-time data processing via Wi-Fi/WebSocket? Would this use case qualify as a "companion device" scenario under iOS background execution policies? What is the best practice for handling this kind of hybrid BLE + WebSocket alerting flow to ensure timely user notifications, even in background/locked/force-closed states? Any advice, documentation links, implementation patterns, or examples from similar companion device apps would be greatly appreciated. Thanks in advance!

I want to implement a feature where a custom notification sound file is downloaded from the server when the app is first launched and stored locally on the device. When a push notification arrives, the stored sound should be played in all app states, including foreground, background, and terminated (killed) state. Does anyone have an idea on how to implement this in iOS? Specifically, I am looking for guidance on: 1)Downloading and storing the sound file securely on the device. 2)Using the locally stored file for push notification sounds. 3)Ensuring the sound plays correctly in all states, including when the app is not running.

Howdy, I'm trying to figure out how to replicate the following behavior for our app: The system is able to ascertain that the Mac equivalent of some iOS app is installed locally, and it prevents notifications from being mirrored. However, I am unable to determine how this association is inferred. When I check our iOS app under this prefpane, the switch remains enabled and toggleable—we'd like to act like Slack here. My initial assumption is that an app group containing both the Mac and iOS apps can be used to create the association; however, I would like to confirm that this is indeed the case before doing so. I'm not terribly confident about this. Details: The bundle identifiers of both apps do not match. This also applies to Slack; its iOS app is com.tinyspeck.chatlyio while its Mac app is com.tinyspeck.slackmacgap. In our case, the iOS app's identifier is like com.company.app while the Mac app's identifier is com.company.app.desktop. Both apps are signed with certificates that have matching team identifiers. The com.apple.developer.team-identifier entitlement is present on the Mac app. The Mac app shares a keychain access group with the iOS app. The Mac app is not sandboxed. The Mac app is an Electron app. The Mac app does not use APNs. It sends notifications "locally". I currently only have the iOS app installed on my iPhone via TestFlight, if that matters. Notification mirroring does work, but we'd like to forcibly disable this by associating the apps together. To my knowledge, the iOS app makes use of both a UNNotificationServiceExtension and a UNNotificationContentExtension. The iOS app currently doesn't have an assigned category (at least in Xcode). The Mac app is currently miscategorized as a developer tool (LSApplicationCategoryType = "public.app-category.developer-tools";), but that should be fixed. (Redacted) bundle information for the Mac app: CFBundleDisplayName = App; CFBundleExecutable = "App Desktop"; CFBundleName = App; Note that our CFBundleExecutable differs from the bundle's display name/name because we're currently migrating our users to a new version of the app that they'd likely want to live alongside the new one. The filename of the bundle itself is, similarly, App Desktop.app. For the iOS app, to my knowledge, the CFBundleName and CFBundleDisplayName are App.

Dear Apple support team, I have some questions for you about this topic: I'm designing an app that need to send a notification like a classic clock alarm timer to the user, let's say each and every 25 minutes of cellphone usage, ofc only if the device is being used. The thing is that when my app is put to background, then Apple will shut it down after 30-60secs. I was thinking to send a silent push notification to my app just so I can wake it up for an instant just so it can shoot a local notification, and then keep in the background totally inactive. That means I'll have to send to my app around 3 silent push notifications x hour. Do you think it is a problem for you? or it's ok to do that? Let me know your experience on this matters, Thanks a lot for your time, A.

Good morning all! We are facing a specific case dealing with push notifications to iOS devices. In my scenario: I turn off my device's internet Send multiple push notifications via server using Firebase. I turned ON my device's internet again. I only see the last push notification I sent. This is an expected scenario? There is any documentation that supports this statement? Thank you all!

Good morning fellow developers, For a while i am struggeling with providing sound to my users on IOS (Safari on Mac is no problem and every other device is not a problem) (we have an existing phone system and made a chat as well), the case is very simple: the notification sound is only for users who are logged in and online for chat. i have tried multiple things: Audio play with javascript (start with mute, play when user clicks a button so the sound is familiar, play when user clicks a button and directly pause it and continue when needed) PWA: the dashboard has been made available as pwa and notifications using google firebase. The popup does show for notifcations to be allowed (and receiving the notifications does work on any other device) But any IOS device cannot register. The information i find is that notifications were supported with 16.4 or higher but also have been deprecated around IOS 17, auto play is not allowed. We have an app in development for our product as well were we will have a notification which will handle this, but that is not the solution we can use now. Long story, short question: is it still somehow possible to push a notification to the user when using the PWA or play a sound in the browser (based on an ajax function). The app/website wont be in the background, so it will always be on the screen. Languages we use: html/javascript (mostly vanilla)/php

I'm strugling about the way how to code notifications for my weather aplication. I use data from my server that receives weather changing values from my own weather station and want to notify user of my app when eg strong wind will blow or temperature go under eg 3℃ etc. The weather station has 8 sensors so there is sometimes a lot of data changing in particular minute that i set to parse data from server and notify user about it. But the notifications only works only when app is on and couple minutes after locking display. So please what could i use strategy for the app to works even when the app sleeps ?

I am trying to retrieve delivered notifications using UNUserNotificationCenter.getDeliveredNotifications(completionHandler:), but I have encountered an issue: Notifications triggered by UNTimeIntervalNotificationTrigger or UNCalendarNotificationTrigger appear in the delivered list. However, notifications triggered by UNLocationNotificationTrigger do not appear in the list. Here is the code I use to fetch delivered notifications: UNUserNotificationCenter.current().getDeliveredNotifications { notifications in for notification in notifications { print("Received notification: \(notification.request.identifier)") } } The notification is scheduled as follows: let center = UNUserNotificationCenter.current() let content = UNMutableNotificationContent() content.title = "Test Notification" content.body = "This is a location-based notification." content.sound = .default let coordinate = CLLocationCoordinate2D(latitude: 37.7749, longitude: -122.4194) // Example coordinates let region = CLCircularRegion(center: coordinate, radius: 100, identifier: "TestRegion") region.notifyOnEntry = true region.notifyOnExit = false let trigger = UNLocationNotificationTrigger(region: region, repeats: false) let request = UNNotificationRequest(identifier: "LocationTest", content: content, trigger: trigger) center.add(request) { error in if let error = error { print("Error adding notification: \(error.localizedDescription)") } } Why does getDeliveredNotifications not return notifications that were triggered using UNLocationNotificationTrigger? How can I retrieve such notifications after they have been delivered?

when I implementation the UNUserNotificationCenterDelegate func userNotificationCenter(_ center: UNUserNotificationCenter, didReceive response: UNNotificationResponse, withCompletionHandler completionHandler: @escaping () -> Void) { var status = "" if (UIApplication.shared.applicationState == .active) { status = "active" } else if (UIApplication.shared.applicationState == .background) { status = "background" } else if (UIApplication.shared.applicationState == .inactive) { status = "inactive" } completionHandler() } I find that UIApplication.shared.applicationState == .background this case can not execute when application is in background。 why applicationState is inactive not background?

Hi, I am developing the Click & Read web add-on for Chromium, Firefox and Safari. We use xcrun safari-web-extension-converter tool to generate the Safari add-on, with up-to-date MacBook MacOS, Xcode et Safari : Sequoia 15.3.2, Safari Version 18.3.1 (20620.2.4.11.6), XCode Version 16.0 (16A242d). We have updated our addon to Manifest v3, having the Background script as Server Worker "background": { "service_worker": "background.js", "type": "module" } self.addEventListener("activate", (event) => { console.info("Service Worker activated", event); event.waitUntil( self.registration.pushManager .subscribe({ userVisibleOnly: true, applicationServerKey: urlBase64ToUint8Array( process.env.VAPID_PUBLIC_KEY ), }) .then(async (subscription) => { console.info("[Service Worker] Extension is subscribed to push"); const { subscription: savedSubscription } = await getLocalStorageKeyData("subscription"); if (savedSubscription) fetchApi({ url: `${API_SERVER_URL}/subscription/remove/${savedSubscription.keys.auth}`, }); // Remove previous subscription from server on addon activate currentBrowser.storage.local.set({ subscription: subscription.toJSON(), }); // Save subscription in local storage currentBrowser.runtime.setUninstallURL( `${API_SERVER_URL}/subscription/remove/${ subscription.toJSON().keys.auth }` ); // Set uninstall URL to remove notification subscription on addon uninstall fetchApi({ url: `${API_SERVER_URL}/subscription`, reqInit: { body: JSON.stringify(subscription.toJSON()), method: "POST", headers: { "Content-Type": "application/json", }, }, }); }) .catch((error) => { console.error("Push subscribe error: ", error); }) // Subscribe to push notifications ); }); When trying to subscribe the addon instance to our Push server, we get this error : Push subscribe error: NotAllowedError: User denied push permission Our NodeJS backend is using the web-push librabry : https://github.com/web-push-libs/web-push) to save subscriptions and make notifications push. By looking for same errors on forums, the best hint I found is that it could be related to the testing is done on localhost (addon is built from XCode onto Safari, and Push server is running on localhost). Thanks for your help !