152

When I try to launch my AndEngine Activity, I get this error:

ERROR/InputDispatcher(21374): channel '4122e148 my.package.AcGame (server)' ~ Channel is unrecoverably broken and will be disposed!

The app doesn't crash, but there's a black screen and the device doesn't react to pressing the 'back' or 'home' buttons.

Does anyone know what the problem is?

Anders R. Bystrup
  • 15,729
  • 10
  • 59
  • 55
uncle Lem
  • 4,954
  • 8
  • 33
  • 53
  • Not much can be guessed from the information you gave (Please add more). But this might help: http://stackoverflow.com/questions/5551929/inputdispatcher-error Check for memory leaks. – Jong Nov 03 '12 at 19:24
  • 17
    That error shows up after an app crashed (or was force-stopped). The part in Android that forwards input events (touchscreen presses etc) to your app has noticed that it's target is no longer there. Look for an error that happens before that one. – zapl Nov 22 '12 at 12:45
  • 1
    @uncle Lem , Bro even i am stuck up in the same issue . I cant perform any operations until i reboot the phone. did u get any solution ? I am fed up with this issue.. – krishnamurthy Oct 12 '18 at 07:23
  • @uncle So did you find the solution? – Gary Chen Oct 16 '20 at 04:16

29 Answers29

53

One of the most common reasons I see that error is when I am trying to display an alert dialog or progress dialog in an activity that is not in the foreground. Like when a background thread that displays a dialog box is running in a paused activity.

Lou Morda
  • 5,078
  • 2
  • 44
  • 49
  • 3
    Set the dialog to null in onPause, and check for null before you display the dialog in the background thread. – Lou Morda Dec 01 '14 at 19:23
  • In my case I got the error because I commented out onResume(), onStart, onStop, onPause, onDestroy, and onLowMemory methods. Thank you @LouMorda for the hint! – Maryoomi1 Jan 10 '19 at 11:03
18

I think that You have memory leaks somewhere. You can find tips to avoid leaking memory here. Also you can learn about tools to track it down here.

Roman Black
  • 3,501
  • 1
  • 22
  • 31
9

Have you used another UI thread? You shouldn't use more than 1 UI thread and make it look like a sandwich. Doing this will cause memory leaks.

I have solved a similar issue 2 days ago...

To keep things short: The main thread can have many UI threads to do multiple works, but if one sub-thread containing a UI thread is inside it, The UI thread may not have finished its work yet while its parent thread has already finished its work, this causes memory leaks.

For example...for Fragment & UI application...this will cause memory leaks.

getActivity().runOnUiThread(new Runnable(){
    public void run() { //No.1          
        ShowDataScreen();

        getActivity().runOnUiThread(new Runnable(){
            public void run() { //No.2
                Toast.makeText(getActivity(), "This is error way",Toast.LENGTH_SHORT).show();
            }});// end of No.2 UI new thread
    }
});// end of No.1 UI new thread

My solution is rearrange as below:

getActivity().runOnUiThread(new Runnable(){
    public void run() {//No.1
        ShowDataScreen();
    }}); // end of No.1 UI new thread       

getActivity().runOnUiThread(new Runnable(){
    public void run() {//No.2
        Toast.makeText(getActivity(), "This is correct way",Toast.LENGTH_SHORT).show();
    }
}); // end of No.2 UI new thread for you reference.

Rohit Gupta
  • 4,022
  • 20
  • 31
  • 41
Jackie_Hung
  • 131
  • 1
  • 4
5

I got similar error (my app crashes) after I renamed something in strings.xml and forgot to modify other files (a preference xml resource file and java code).

IDE (android studio) didn't showed any errors. But, after I repaired my xml files and java code, app ran okay. So, maybe there are some small mistakes in your xml files or constants.

Rohit Sharma
  • 1,271
  • 5
  • 19
  • 37
ishitcno1
  • 713
  • 9
  • 10
5

You can see the source code about this output here:

void InputDispatcher::onDispatchCycleBrokenLocked(
        nsecs_t currentTime, const sp<Connection>& connection) {
    ALOGE("channel '%s' ~ Channel is unrecoverably broken and will be disposed!",
            connection->getInputChannelName());
    CommandEntry* commandEntry = postCommandLocked(
            & InputDispatcher::doNotifyInputChannelBrokenLockedInterruptible);
    commandEntry->connection = connection;
}

It's cause by cycle broken locked...

codezjx
  • 9,012
  • 5
  • 47
  • 57
5

I had the same problem. To solve the error: Close it on the emulator and then run it using Android Studio.

The error happens when you try to re-run the app when the app is already running on the emulator.

Basically the error says - "I don't have the existing channel anymore and disposing the already established connection" as you have run the app from Android Studio again.

Frames Catherine White
  • 27,368
  • 21
  • 87
  • 137
Kapil Bhagia
  • 65
  • 1
  • 7
2

I had the same problem but mine was Due To an Android database memory leak. I skipped a cursor. So the device crashes so as to fix that memory leak. If you are working with the Android database check if you skipped a cursor while retrieving from the database

amadosi
  • 21
  • 3
  • I didn't used no databases in that app. Seems to me that there's a LOT of options that may cause that error. – uncle Lem Aug 27 '13 at 13:33
2

I had the same problem. Mine was due to a third jar, but the logcat did not catch the exception, i solved by update the third jar, hope these will help.

yangzx
  • 51
  • 2
2

As I faced this error, somewhere in your code your funcs or libraries that used run on different threads, so try to call all code on the same thread , it fixed my problem.

If you call methods on WebView from any thread other than your app's UI thread, it can cause unexpected results. For example, if your app uses multiple threads, you can use the runOnUiThread() method to ensure your code executes on the UI thread:

Google reference link

Hamed Jaliliani
  • 2,789
  • 24
  • 31
2

It's obvious this creeps up due to many issues. For me, I was posting several OneTimeWorkRequest, each accessing a single room database, and inserting into a single table.

Making the DAO functions suspended, and calling them within the coroutine scope of the worker fixed this for me.

  • 1
    Could you say more what you mean with "Calling them within the coroutine scope of the worker fixed this"? I think I have the same issue but I am not able to fix it – Andrew Dec 11 '20 at 21:37
  • @Andrew If the DAO functions are suspended, then they have to be called from some coroutine scope. For my case, the database updates were being performed inside a `OneTimeWorkRequest` which as an Android Worker, has an extension library which surfaces the coroutine scope for the worker. Even more specific, these requests were appended to one-another, so with many requests at one time, I'm not sure if coroutines alone will handle queuing updates to the database. In this case (without a worker that is queuing requests), it might be worth looking into Kotlin's Mutex syntax. – Matthew Wood Dec 17 '20 at 12:58
1

I was having the same problem too. In my case was caused when trying to reproduce videos with a poor codification (demanded too much memory). This helped me to catch the error and request another version of the same video. https://stackoverflow.com/a/11986400/2508527

Community
  • 1
  • 1
1

It happened for me as well while running a game using and-engine. It was fixed after i added the below code to my manifest.xml. This code should be added to your mainactivity.

android:configChanges="keyboard|keyboardHidden|orientation|screenLayout|uiMode|screenSize|smallestScreenSize|mcc|mnc"
mallik
  • 557
  • 4
  • 13
1

Reading through all contributions, it looks like many different origins exhibit cause this same problem symptoms.

In my case for instance - I got this problem as soon as I added

android:progressBackgroundTintMode="src_over"

to my progress bar properties. I think the GUI designer of ADT is known for several bugs. Hence I assume this is one of them. So if you encounter similar problem symptoms (that just do not make sense) after playing with your GUI setup, just try to roll back what you did and undo your last GUI modifications.

Just press Ctrl+z with the recently modified file on screen.

Or:

The Version Control tool could be helpful. Open the Version Control panel - choose Local Changes tab and see recently modified (perhaps .xml) files.

Right click some most suspicious one and click Show Diff. Then just guess which modified line could be responsible.

Good luck :)

Sold Out
  • 1,321
  • 14
  • 34
1

In my case these two issue occurs in some cases like when I am trying to display the progress dialog in an activity that is not in the foreground. So, I dismiss the progress dialog in onPause of the activity lifecycle. And the issue is resolved.

Cannot start this animator on a detached view! reveal effect BUG

ANSWER: Cannot start this animator on a detached view! reveal effect

Why I am Getting Error 'Channel is unrecoverably broken and will be disposed!

ANSWER: Why I am Getting Error 'Channel is unrecoverably broken and will be disposed!'

@Override
protected void onPause() {
    super.onPause();
    dismissProgressDialog();

}

private void dismissProgressDialog() {
    if(progressDialog != null && progressDialog.isShowing())
        progressDialog.dismiss();
}
jazzbpn
  • 6,441
  • 16
  • 63
  • 99
1

I had this issue and the cause was actually a NullPointerException. But it was not presented to me as one!

my Output: screen was stuck for a very long period and ANR

My State : the layout xml file was included another layout, but referenced the included view without giving id in the attached layout. (i had two more similar implementations of the same child view, so the resource id was created with the given name)

Note : it was a Custom Dialog layout, so checking dialogs first may help a bit

Conclusion : There is some memory leak happened on searching the id of the child view.

Blue_Alien
  • 2,148
  • 2
  • 25
  • 29
1

This error occurred in case of memory leak. For example if you have any static context of an Android component (Activity/service/etc) and its gets killed by system.

Example: Music player controls in notification area. Use a foreground service and set actions in the notification channel via PendingIntent like below.

Intent notificationIntent = new Intent(this, MainActivity.class);
        notificationIntent.setAction(AppConstants.ACTION.MAIN_ACTION);
        notificationIntent.setFlags(Intent.FLAG_ACTIVITY_NEW_TASK
                | Intent.FLAG_ACTIVITY_CLEAR_TASK);
        PendingIntent pendingIntent = PendingIntent.getActivity(this, 0,
                notificationIntent, 0);

        Intent previousIntent = new Intent(this, ForegroundService.class);
        previousIntent.setAction(AppConstants.ACTION.PREV_ACTION);
        PendingIntent ppreviousIntent = PendingIntent.getService(this, 0,
                previousIntent, 0);

        Intent playIntent = new Intent(this, ForegroundService.class);
        playIntent.setAction(AppConstants.ACTION.PLAY_ACTION);
        PendingIntent pplayIntent = PendingIntent.getService(this, 0,
                playIntent, 0);

        Intent nextIntent = new Intent(this, ForegroundService.class);
        nextIntent.setAction(AppConstants.ACTION.NEXT_ACTION);

        Bitmap icon = BitmapFactory.decodeResource(getResources(),
                R.drawable.ic_launcher);

        NotificationManager notificationManager = (NotificationManager) getSystemService(Context.NOTIFICATION_SERVICE);
        String NOTIFICATION_CHANNEL_ID = "my_channel_id_01";

        if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.O) {
            NotificationChannel notificationChannel = new NotificationChannel(NOTIFICATION_CHANNEL_ID, "My Notifications", NotificationManager.IMPORTANCE_HIGH);

            // Configure the notification channel.
            notificationChannel.setDescription("Channel description");
            notificationChannel.enableLights(true);
            notificationChannel.setLightColor(Color.RED);
            notificationChannel.setLockscreenVisibility(Notification.VISIBILITY_PRIVATE);
            notificationChannel.setVibrationPattern(new long[]{0, 1000, 500, 1000});
            notificationChannel.enableVibration(true);
            notificationManager.createNotificationChannel(notificationChannel);
        }

        NotificationCompat.Builder notificationBuilder = new NotificationCompat.Builder(this, NOTIFICATION_CHANNEL_ID);
        Notification notification = notificationBuilder
                .setOngoing(true)
                .setAutoCancel(true)
                .setWhen(System.currentTimeMillis())
                .setContentTitle("Foreground Service")
                .setContentText("Foreground Service Running")
                .setSmallIcon(R.drawable.ic_launcher)
                .setLargeIcon(Bitmap.createScaledBitmap(icon, 128, 128, false))
                .setContentIntent(pendingIntent)
                .setPriority(NotificationManager.IMPORTANCE_MAX)
                .setCategory(Notification.CATEGORY_SERVICE)
                .setTicker("Hearty365")
                .build();
        startForeground(AppConstants.NOTIFICATION_ID.FOREGROUND_SERVICE,
                notification);

And if this notification channel get broken abruptly (may be by system, like in Xiomi devices when we clean out the background apps), then due to memory leaks this error is thrown by system.

Akki
  • 775
  • 8
  • 19
  • 1
    Hi, I think this could be my problem. I see this error every time I close my media player. I think I clean everything correctly, I hide the notification, unregister receivers etc. But I get this message anyway. The application behavior is not affected by it, but I would like to fix it anyway. What should I do to avoid it? – JeCh Nov 14 '19 at 09:05
  • @JeCh found anything? – fsljfke Aug 30 '20 at 22:28
1

For me it was caused by a splash screen image that was too big (over 4000x2000). The problem disappeared after reducing its dimensions.

Dominique Lorre
  • 1,168
  • 1
  • 10
  • 19
1

Just Try to Invalidate the IDE cached and restart. This wont fix the issue. But in my case doing this revealed the possible crash

RAINA
  • 802
  • 11
  • 22
1

Please check you Realm entity class.

If you declare variable as lateinit var and you try to check that uninitialized variable check with isNullOrEmpty() return "Channel is unrecoverably broken and will be disposed!"

Ole Pannier
  • 3,208
  • 9
  • 22
  • 33
1

In my case I was setting value in background thread viewModelScope.launch(Dispatchers.IO) { } so app was crashing solution in my case I was removed (Dispatchers.IO). viewModelScope.launch{ }

  • Yes. Completely true. This strange "server-channel error" is related to changing Views from another thread. In my case I just wrapped that calls with Handler(Looper.getMainLooper()).post { /*change ui here*/ } – Eugene P. Aug 24 '22 at 11:31
0

In my case, I was using Glide library and the image passed to it was null. So it was throwing this error. I put a check like this:

if (imageData != null) {
    // add value in View here 
}

And it worked fine. Hope this helps someone.

Adnan Afzal
  • 167
  • 1
  • 3
  • 8
0

I got same logcat message, just realize that string.xml value of array cannot be number/digit, but only text/alphabet is allowed.

LBIRD S.E.
  • 31
  • 3
0

In my case this error is happening because of not connected to firebase firestore but using the same.

To rectify the issue please go to Tools-> Firebase

when a window will open on RHS choose the options to -> Connect your app to Firebase -> Add Cloud Firestore to your app

Christopher Moore
  • 15,626
  • 10
  • 42
  • 52
0

Check your ViewModel class and not finding any issue then try to comment code where you using launch or withContext.

In my case, I commented on code where I am using launch or withContext and it worked my app is running normally.

iamkdblue
  • 3,448
  • 2
  • 25
  • 43
0

I had a kind of issue you described. In my case it happened only in release builds. It was caused by the obfuscation: native methods crashed silently on FindClass or GetMethodID call cuz the names were obfuscated. Editing proguard-rules worked out!

ololo
  • 1
  • This does not really answer the question. If you have a different question, you can ask it by clicking [Ask Question](https://stackoverflow.com/questions/ask). To get notified when this question gets new answers, you can [follow this question](https://meta.stackexchange.com/q/345661). Once you have enough [reputation](https://stackoverflow.com/help/whats-reputation), you can also [add a bounty](https://stackoverflow.com/help/privileges/set-bounties) to draw more attention to this question. - [From Review](/review/late-answers/29951834) – Sangeerththan Balachandran Sep 29 '21 at 21:23
0

The issue for me was I havent defined the instance of an Activity.

Eg:

Myclass my;

onCreate() {

my.getData(); }

Instead of:

Myclass my;

onCreate() {

my = new Myclass(); my.getData(); }

This is weird as Studio has to give some good defining error message.

Prajwal Waingankar
  • 2,534
  • 2
  • 13
  • 20
0

Just to add to many other answers:

My app crashed without anything relevant in LogCat. My Retrofit network call that caused the crash looked like this:

CoroutineScope(Dispatchers.IO).launch {
   val response = api.getData()
}

and used the following method signature

suspend fun getData() : Response<RatesResponse>

I changed it removing Response from the return type & got it working

suspend fun getData() : RatesResponse
0

There are two ways this can be resolved,

  1. Try removing the Dispatchers.XXX from the launch

    viewModelScope.launch(Dispatchers.IO) { } viewModelScope.launch{ }

  2. Add CoroutineExceptionHandler to catch the exception, this will tell you what exactly is going wrong.

    val handler = CoroutineExceptionHandler { coroutineContext, throwable -> Log.d("TAG","EROR ${throwable.message}") } lifecycleScope.launch(handler) { }

Raghav Sharma
  • 43
  • 1
  • 6
0

When it happened to me:

  1. Forgot to delete TODO().
  2. SQL query result with a null field getting assigned to a non null data class's field.
Taras Mykhalchuk
  • 829
  • 1
  • 9
  • 20