We have a video chat single page application developed in AngularJS that connects Callers with Receivers using OpenTok (Tokbox). Receivers must use computers but Callers can use iOS or Android devices. However, when a Caller uses iOS, the first call is fine, but the 2nd and subsequent calls in the same session result in audio being lost for the iPhone user.
The application consists of a Homepage, Profile pages, Dashboards and Call Screens. A Caller initiates a video chat with a Receiver in the following way:
Types in Receiver’s Profile page and logs in
Clicks ‘Call’ from Receiver’sProfile page, lands on Call Screen & gives permission to use camera etc.
Clicks ‘Call’ again and video/audio chat starts
Clicks ‘End call’ and lands on their dashboard page
Then, if the Caller chooses to call Receiver again, the Caller:
Either clicks back button or retypes Receiver’s URL to return to Receiver’s profile page
Clicks ‘Call’ from Receiver’s Profile page, lands on call screen
Clicks ‘Call’ again and video starts - but usually (although not always) cannot hear any audio on the iPhone
After extensive testing, only 3 things happen with 100% consistency:
If, after a chat, the Caller goes to the Homepage and then returns to Profile page, the audio will never work, no matter how many times you try.
If, however, the Caller types in a URL of an entirely different site, and returns to the profile page, the problem is always reset: the Caller will definitely have audio for the next call, after which the whole problem starts again.
Closing and initiating a new window also resets the issue.
Aside from these certainties, there are many other variables that may or may not produce the issue (and how often), including:
What version of iOS the Caller is using
Whether the Caller uses the back button or types in the URL to go from dashboard to profile page after a call
Whether the Caller simply clicks the back button or holds and chooses the Profile page URL
Whether the Caller clicks around in their Dashboard or not before returning to Profile page
Tokbox session logs show that both published streams contain both audio and video, and no errors are reported. The audio is lost within iOS.
We have no problems with computer to computer, or Android to computer: Only iOS.
We have attempted almost everything to solve this issue - from stripping back almost all 3rd party APIs to removing most of the AngularJS code and nothing solves this mysterious (and very annoying) problem.
Versions: Angular 1.5, Opentok: 2.16.2
Please let me know if anybody came across anything like this.
Related
Background:
We are building a React Native app which allows users to create posts that include a quote from a blog post or news article. To achieve this, we have built a share extension, which displays a React Native view inside a modal. Inside the RN view we are using react-native-webview (which renders as a WKWebView on iOS), to display the page that was shared, allowing the user to select some text on the page (which is fed back to our app via an injected content script), and continue to create a post.
The problem:
Some news sites that our users want to share from have paywalls, e.g. FT, NY Times etc. - on these sites the user has to log in inside the webview. It's not ideal, but once they have logged in, as we have cookies enabled on the webview, their session will be remembered. This solution works well on Android, and works well on iOS when sharing from Safari.
When sharing from an app on iOS however, once the user submits the login form inside the webview, the share extension closes, taking the user back to the app that initiated the share action.
Here is a video demonstration of the issue occurring in the Guardian app:
https://www.youtube.com/watch?v=DFnh1x3j4xs
Observations:
When observing requests that occur within the webview using the onShouldStartLoadWithRequest prop of the WebView component, upon submitting the login form, there are two requests where iOS reports the navigationType as formSubmitted.
By checking the navigationType inside of the onShouldStartLoadWithRequest method, I was able to allow the first formSubmitted request to go through, but cancel the subsequent one. This prevented the share extension from closing, but upon clicking a link to go back to the home page, the share extension was closed.
My current theory is that the app which initiates the share, is somehow hooking into the form submit event, as these apps generally use webviews for their own login process - so although this webview is running in our own process, perhaps the app still somehow hooks into the process.
It is worth noting that this behaviour is inconsistent between news apps, presumably due to differences in the way they handle their login process.
Initially this issue was occurring in the BBC app as well, but at some point stopped occurring without us making any changes, so it is somewhat intermittent, or perhaps based on stored data/cookies.
Summary
We would like to fully understand why this behaviour occurs, and we're hoping there is a solution or workaround that will prevent the share extension from closing when the login form is submitted.
I'm working on two iOS applications that share the same keychain tokens which they receive when the user logs-in in either of the apps by entering his username and password in a web view. The tokens saved saved in the keychain, are time limited for a few hours/days - they are being refreshed when the user uses the app and makes request/enters items which are being opened in a browser.
Up until the release of iOS 11, the web view in the app was a SFSafariViewController instance and it was very convenient because it shared the cookies between the apps and once a user logged-in in a browser in one app, he would automatically be logged-in in the other app, thus achieving an SSO experience.
In iOS 11, the behavior of SFSafariViewController changed, and it no longer shares the cookies between different SFSafariViewController instances in my two apps. Instead, Apple wants us to use SFAuthenticationSession to achieve the same behavior.
In my app, I have a table of about 50 rows, almost all of them open a different URL in SFSafariViewController. So now, in iOS 11, I need to create a new instance of SFAuthenticationSession every time (the URL cannot be changed once the instance was created), and in order to actually show the web view, I need to invoke the start() method on this instance. However, this method can only be invoked once on the same instance (otherwise it will return 'false' and do nothing), and every time it is being invoked, it pops the notification that "myapp" wants to use "mydomain.com" to sign in.
So it means that my users see this alert message on every tap on a row in the table.
Is there a way to make my app show the consent alert message only once, e.g. on the first time or something? I want my users to have an SSO experience in my two apps, that's why I used SFSafariViewController in the first place, but I don't want them to see this annoying alert on every tap they make in the app.
There is currently no way of having the alert message only show once using SFAuthenticationSession, we are experiencing the same issue on our apps.
I along with many others have filed a bug regarding this to Apple, asking them to show it only once like you suggested. The end of this GitHub thread discusses the alert.
I have an iOS app that permits people to "claim points." I want to prevent a scenario where someone opens the app on their phone, another instance on an iPad and then submits claims in both places.
Presently the app "registers" as the active instance when it starts up. If a second instance starts, it will take control and any attempt by the first instance to save data will result in a redirect that tells the user not to have two apps open. I have seen some games (e.g. Clash) that immediately inform the user of the first app that they are being kicked the moment that the second app opens.
I can think of two ways to do this: submit a "heartbeat" every few seconds to see if we are still the controlling app (wasteful) and some form of notification to all other apps when a new instance opens. Can standard iOS notifications play this role or are they just for popping up notifications to the user?
Any insight into how to post a notice to all other instances of an app when one instance opens up that doesn't require polling would be appreciated!
You can manage this problem on the server side.
Apparently users have to login or at least, your app auto-login with a unique ID with a server.
If you have an API to receive a claim, you can simply check if an identical claim already exists for this Unique ID and return an error to the second device when that is the case.
I created a router that connects to facebook to get some info before a user may access the internet.
First they connect, get the Captive Portal Page and then continue to a facebook login. Since the upgrade to iOS7 it fails to load the facebook login page. On my mac with the Captive Portal Assistant it has no problems and even on the phone itself while using the iOS version of safari there are no problems.
What is going wrong here? Is facebook filtering request from the iOS7 Captive Portal Assistant or is Apple doing some sneaky stuff here?
It seems the problem is widespread and only related to facebook.
Update: I worked with the beta's and they worked fine a few weeks ago. Now with the same beta version it doesn't anymore. So another point for the facebook explaination.
Regards, Cas
This problem was fixed by Apple since IOS 8. But as all iPhone 4 users can't upgrade to IOS 8 this problem is still one.
The IOS 7 devices check for the following domains:
www.appleiphonecell.com
captive.apple.com
captive.apple.com
www.apple.com
www.itools.info
www.ibook.info
www.airport.us
www.thinkdifferent.us
Whitelisting this domains stops the login mask to be appearing as the IOS device thinks, that the internet is working as expected. This way you have control on the things which happens, as the IOS device does not interrupts anything, if you use a normal browser for login.
If you don't whitelist the domains, the following thing happens. I debugged it on routers with several IOS devices and they all did mostly the same:
If you connect to a wifi, the IOS device tries to connect to one of the domains, which are listed above. If it can contact one of the domains, it tries another one. If it can't, it starts the redirect, which is controlled by the router. Sometimes it query one or more domains, before it thinks, that the internet is working.
After the check, the login screen redirects to your router and then to the login screen of your captive portal. This behavior stayed the same as it was on IOS 6 or before.
Now you start an oauth login to a 3rd party provider like Facebook, Google or Twitter. And now the difference appears. You can check it on a router, if you run it in debug mode. As the IOS device goes to an other domain for oauth login (like www.facebook.com) the iPhone thinks, that something changed and starts to query one of the apple domains, which are listed above. The user sees only a white screen and in the background the IOS device tries to contact one of the domains repeatedly. For the user this seems to be an error, as the screen stays white or takes very long to show the login on the 3rd party provider. Sometimes it stops loading and nothing happens forever.
To avoid this behavior, you must whitelist the above listed domains. This is a not a common behavior for IOS users, but this way, your browser have the control of the login session and the IOS device don't interrupt it as it does with the login screen.
Some shallow parts of information is reported on the following sites:
https://supportforums.cisco.com/docs/DOC-36523
http://www.cadincweb.com/why-your-apple-ios-7-device-wont-connect-to-the-wifi-network
https://discussions.apple.com/thread/5355766
I couldn't find a detailed description of the problem and found the one above myself by debugging all parts with some routers and IOS devices like iPhones and iPads.
I've just tested various router settings and noticed that iOS 7 is NOT trying to contact above mentioned sites/URLs when router's DOMAIN field is blank.
My guess is that blank domain points to a consumer-type network set up and Apple is not expecting a Captive Portal at such network. If you have access to administer your router see if you can clear out the DOMAIN field (and restart/retest).
I found my solution to my problem. (a while ago, but I found this post again)
First I found out, iOS makes 3 calls, first to check, second to get the page that needs to be displayed, third to check again after the pageload. Then I discovered, for every POST or GET action made by the page, regarding of the source page was refreshed, iOS checks for an active internet connection. Since the facebook api makes a lot of calls, the browser starts stalling(possible in combination with QoS on my router) and freezes the page.
My Solution:
Since I am in control of the DNS records of the Router I use, I redirected all domains towards my own server.
First I saved the check request, this to later identify the user when he comes back for the 3th request.
When the second request comes I just display an info window that every thing is right, and the user has to click the "Done" button.
The page is loaded, so iOS checks again, but i recognize the user so I display the OK-code Apple also displays. The "Done" button us show, and the user has "internet", according to iOS..
On the page that I display, I instruct the user to open the webbrowser. When he does, he opens a page and my portal with the right page is shown(I can detect this based on the Browser Agent). Then my facebook api start doing its job, and of we go :-)
Let me know if someone needs some more info on how to detect or maybe even a code sample if necessary.
Extra Information
To capture a user on your own server, redirect every request to your processing page using for example .htaccess. The request is made to a domain with a subfile e.g:
http://captive.apple.com/getrT09Nx7G/YNrnUOulnDj/3cfrq3M40iR.html
To keep multiple users apart, use the unique url the device tries to contact when checking for internet, in this case: /getYT09Nx7G/YN1nUOulnDj/3cfMq3M40iR.html
I've got (actually my employer has) a mobile website that enables Safari integration (for iPhones and iPads) - meaning that customers can bookmark it to their home screen and then it would behave as a standalone web app (no address bar, custom icon, start-up image etc).
It works all right except that one week ago (coincidentally soon after apple has released iOS 6.1.2) some of our customers (6 of them initially) complained that they no longer get the normal content but a '404 page' of a public wifi provider (The Cloud owned by Sky here in the UK). After a bit of investigation we've figured that at some point those customers connected to the Cloud wifi without actually logging in (it's one of those providers that would redirect you to a login page to enter your credentials, after which you can carry on browsing). The thing is that even after switching back to their private wifi or mobile data connection the application would display the Cloud's page.
This only happens (as far as I can tell) when the application is launched via the bookmark (I couldn't see this behavior when using it from safari).
What happens is that the customers would connect to the cloud wifi (without logging in), they would open the application at which point the router will issue a redirect response to their login page; the application would cache the login page and it will always display it whenever using the bookmark again. (I've performed a capture when this happens and there are no requests being made at start-up whatsoever).
Even weirder, in this situation, if removing the existing bookmark and adding a new one will show you the same cached page (with the whole operation being performed away from the Cloud). We've fixed this by adding a unique identifier to the URL each time we hit the bookmark screen (this indicates that the web apps' sandboxes are linked to the url, which is to be expected).
What we're trying to achieve is to have the application properly recovering after the customer has moved away from the Cloud. But there doesn't seem to be a straight forward way to do this.
Furthermore there's a level of inconsistency in all of this - most of the times when the flow is performed I will see a 404 page (a custom 404 page https://service.thecloud.net/service-platform), but sometimes I would be properly redirected to the login page, in which case the application would not break.
My assumption is that there is a weird race condition in the standalone web app application model causing the browser not to properly handle redirects (and actually caching 404 pages). I've raised a support incident with Apple (which eventually turned into a bug report) but it might take a while and I'm trying my best to figure out a workaround.
Any ideas, maybe someone has seen this before?
The issue is aggravated by the fact that I need to have a 5 minutes walk ever time I'm testing any fixes; I've tried creating simple test forms, but I wasn't able to reproduce the issue, where as with the full app I can do it pretty much every time.
Here's a summary of the steps to reproduce:
Via private wifi (or mobile data connection) add a bookmark to a website (I've managed to reproduce it with quite a couple of apps that support safari integration as described above)
Open the application to review the normal content
Connect to a Cloud hotspot and open the application from the bookmark (open-close it for a couple of times if you don't get the 404 right away)
Connect to the private wifi (or mobile data connection) and open the application via the bookmark -> you'll see the same 404 page again
In the end the fix was to add a unique query string parameter with the initial page request (pretty easy with the setup we already had, via the launcher page). I've filed a bug report with Apple which they've acknowledged by linking it to a previous item. Here's a post on the topic:
http://blog.onos.ro/ios-6.1.2-caching-issue