IssueTracker

I guess technically, the ProGuard rule should be on androidx.work.NonBlockingWorker subclasses, but I'd also be okay with just not using reflection to access the Worker(Context, WorkerParamters) constructor and skipping the need for a ProGuard rule at all.

Fixed in https://android-review.googlesource.com/c/platform/frameworks/support/+/767262

The fix will be available in alpha10.

I tried to use the ProGuard configuration which is linked in comment #3 in my project.
The release build fails and the following warnings are printed:

22:22:06.567 [QUIET] [system.out] Note: the configuration keeps the entry point 'androidx.work.NonBlockingWorker { NonBlockingWorker(android.content.Context,androidx.work.WorkerParameters); }', but not the descriptor class 'androidx.work.WorkerParameters'
22:22:06.567 [QUIET] [system.out] Note: the configuration keeps the entry point 'androidx.work.NonBlockingWorker { void internalInit(android.content.Context,androidx.work.WorkerParameters); }', but not the descriptor class 'androidx.work.WorkerParameters'
22:22:06.567 [QUIET] [system.out] Note: the configuration keeps the entry point 'androidx.work.Worker { Worker(android.content.Context,androidx.work.WorkerParameters); }', but not the descriptor class 'androidx.work.WorkerParameters'

The app is opensource if you want to run the release build yourself. Here is the branch (WIP):
https://github.com/EventFahrplan/EventFahrplan/tree/workmanager
Please run: ./gradlew --info --debug aCcc34c3Release

Ah, that's because we have a typo in our Proguard file.  Change WorkParameters to WorkerParameters.

Thank you for quick reply. I changed this here:
https://github.com/EventFahrplan/EventFahrplan/commit/7567331dc69a303d5c423850a97bf29005ea7b1c

The release build still fails though. Here is a part of the log output:

23:43:15.760 [ERROR] [system.err] Note: there were 11 duplicate class definitions.
23:43:15.760 [ERROR] [system.err]       (http://proguard.sourceforge.net/manual/troubleshooting.html#duplicateclass)
23:43:15.822 [QUIET] [system.out] Initializing...
23:43:15.827 [QUIET] [system.out] Note: the configuration refers to the unknown class 'androidx.work.impl.background.firebase.FirebaseJobScheduler'
23:43:15.845 [QUIET] [system.out] Note: the configuration refers to the unknown class 'androidx.work.WorkParameters'
23:43:15.857 [QUIET] [system.out] Note: the configuration refers to the unknown class 'androidx.work.impl.background.firebase.FirebaseJobScheduler'
23:43:16.272 [ERROR] [system.err] Warning: androidx.concurrent.futures.AbstractFuture$UnsafeAtomicHelper: can't find referenced class sun.misc.Unsafe
23:43:16.272 [ERROR] [system.err] Warning: androidx.concurrent.futures.AbstractFuture$UnsafeAtomicHelper: can't find referenced class sun.misc.Unsafe
23:43:16.272 [ERROR] [system.err] Warning: androidx.concurrent.futures.AbstractFuture$UnsafeAtomicHelper: can't find referenced class sun.misc.Unsafe
23:43:16.272 [ERROR] [system.err] Warning: androidx.concurrent.futures.AbstractFuture$UnsafeAtomicHelper: can't find referenced class sun.misc.Unsafe
23:43:16.272 [ERROR] [system.err] Warning: androidx.concurrent.futures.AbstractFuture$UnsafeAtomicHelper: can't find referenced class sun.misc.Unsafe
23:43:16.272 [ERROR] [system.err] Warning: androidx.concurrent.futures.AbstractFuture$UnsafeAtomicHelper: can't find referenced class sun.misc.Unsafe
23:43:16.272 [ERROR] [system.err] Warning: androidx.concurrent.futures.AbstractFuture$UnsafeAtomicHelper: can't find referenced class sun.misc.Unsafe
23:43:16.272 [ERROR] [system.err] Warning: androidx.concurrent.futures.AbstractFuture$UnsafeAtomicHelper$1: can't find referenced class sun.misc.Unsafe
23:43:16.273 [ERROR] [system.err] Warning: androidx.concurrent.futures.AbstractFuture$UnsafeAtomicHelper$1: can't find referenced class sun.misc.Unsafe
23:43:16.273 [ERROR] [system.err] Warning: androidx.concurrent.futures.AbstractFuture$UnsafeAtomicHelper$1: can't find referenced class sun.misc.Unsafe
23:43:16.273 [ERROR] [system.err] Warning: androidx.concurrent.futures.AbstractFuture$UnsafeAtomicHelper$1: can't find referenced class sun.misc.Unsafe
23:43:16.285 [ERROR] [system.err] Warning: androidx.work.impl.background.systemjob.SystemJobService: can't find referenced method 'android.net.Network getNetwork()' in library class android.app.job.JobParameters
23:43:16.883 [QUIET] [system.out] Note: androidx.work.impl.Schedulers: can't find dynamically referenced class androidx.work.impl.background.firebase.FirebaseJobService
23:43:16.883 [QUIET] [system.out] Note: androidx.work.impl.Schedulers: can't find dynamically referenced class androidx.work.impl.background.firebase.FirebaseJobScheduler
23:43:16.911 [QUIET] [system.out] Note: kotlin.internal.PlatformImplementationsKt: can't find dynamically referenced class kotlin.internal.jdk8.JDK8PlatformImplementations
23:43:16.911 [QUIET] [system.out] Note: kotlin.internal.PlatformImplementationsKt: can't find dynamically referenced class kotlin.internal.JRE8PlatformImplementations
23:43:16.912 [QUIET] [system.out] Note: kotlin.internal.PlatformImplementationsKt: can't find dynamically referenced class kotlin.internal.jdk7.JDK7PlatformImplementations
23:43:16.912 [QUIET] [system.out] Note: kotlin.internal.PlatformImplementationsKt: can't find dynamically referenced class kotlin.internal.JRE7PlatformImplementations
23:43:16.914 [QUIET] [system.out] Note: kotlin.jvm.internal.Reflection: can't find dynamically referenced class kotlin.reflect.jvm.internal.ReflectionFactoryImpl
23:43:16.944 [QUIET] [system.out] Note: okhttp3.internal.platform.AndroidPlatform: can't find dynamically referenced class com.android.org.conscrypt.SSLParametersImpl
23:43:16.944 [QUIET] [system.out] Note: okhttp3.internal.platform.AndroidPlatform: can't find dynamically referenced class org.apache.harmony.xnet.provider.jsse.SSLParametersImpl
23:43:16.944 [QUIET] [system.out] Note: okhttp3.internal.platform.AndroidPlatform$CloseGuard: can't find dynamically referenced class dalvik.system.CloseGuard
23:43:16.944 [QUIET] [system.out] Note: okhttp3.internal.platform.ConscryptPlatform: can't find dynamically referenced class org.conscrypt.ConscryptEngineSocket
23:43:16.944 [QUIET] [system.out] Note: okhttp3.internal.platform.Platform: can't find dynamically referenced class sun.security.ssl.SSLContextImpl

I think your compileSdk is too low.  WorkManager requires it to be 28 or higher: https://irina-dot-devsite.googleplex.com/topic/libraries/architecture/adding-components

I'm not sure why you are running into the warnings about Unsafe, but you can try using -dontwarn sun.misc.Unsafe in your proguard file to see if it helps.  We haven't observed it ourselves.

For what it's worth, one of the notes is referring to WorkParameters still - are you sure it was changed properly?

Sorry, the right url above should be: https://developer.android.com/jetpack/docs/release-notes

I changed the ProGuard configuration as you can see it in the GitHub repository. It matches with your change here https://android-review.googlesource.com/c/platform/frameworks/support/+/782087/2/work/workmanager/proguard-rules.pro if I don't miss anything. I cannot tell why "WorkParameters" is still listed. I verified the output once again by running ./gradlew clean --info --debug aCcc34c3Release. Feel free to run the build yourself. All you need to do is to create a random release-keystore and the app/gradle.properties from the app/gradle.properties.example.

So you saying when I want to migrate my app to targetSdkVersion 26 until November I need to raise the compileSdkVersion to 28 in order to use WorkManager? Using targetSdkVersion 26 with compileSdkVersion 28 won't work?

There shouldn't be any issues with targetSdkVersion 26 and compileSdkVersion 28.

Okay. I will try that. Thanks.
Feel free to contact me if you are not able to build my project.

Does it suffice to raise the compileSdkVersion to 28 or do I also have to raise the version of the support library artifacts at the same time? I tried your suggestion in comment #11 but the build still fails.

Project: platform/frameworks/support
Branch: androidx-master-dev

commit cee9ab55352f3ad506552afb4753baf8648b4c30
Author: Sumir Kataria <sumir@google.com>
Date:   Mon Oct 08 14:20:57 2018

    Fix class name in Proguard configuration.
   
    Bug: 116296569
    Test: N/A
    Change-Id: I3525f9d2f3b06523cbe842b35f97014677879a64

M       work/workmanager/proguard-rules.pro

https://android-review.googlesource.com/782087
https://goto.google.com/android-sha1/cee9ab55352f3ad506552afb4753baf8648b4c30

Support library as well.

Do the latest changes (WorkParameters -> WorkerParameters, NonBlockingWorker -> ListenableWorker) in the ProGuard configuration apply to WorkManager 1.0.0-alpha09 or 1.0.0-alpha10 or both?

alpha10.

Thank you. Can you please point me to the commit/file which represents the configuration for WorkManager 1.0.0-alpha09? Maybe here https://android.googlesource.com/platform/frameworks/support/+log/androidx-master-dev/work/workmanager/proguard-rules.pro?

There isn't one for alpha09... it's already released.  You can modify that commit and adapt it to alpha09.

Just change WorkParameters -> WorkerParameters.

Which commit do you refer to? The link I posted show the history of the proguard-rules.pro commits. Since the file exists for a year I would have expected that there was a version of the file which shipped with WorkManager 1.0.0-alpha which was released by September 19, 2018.

Or do you mean that the proguard-rules.pro being shipped with WorkManager 1.0.0-alpha is buggy?

Yes, alpha09 had a bug.  #19 suggests a modification to your app's proguard config to get around the issue.  alpha10 should be out soon with the fix.

Thank you for the clarification.

I prepared my test branch according to your recommendations (ProGuard, compileSdkVersion, targetSdkVersion, support libraries) as you can see here:
https://github.com/EventFahrplan/EventFahrplan/compare/master...workmanager

The release build however still fails:

00:20:09.182 [QUIET] [system.out] Note: androidx.work.impl.Schedulers: can't find dynamically referenced class androidx.work.impl.background.firebase.FirebaseJobService
00:20:09.182 [QUIET] [system.out] Note: androidx.work.impl.Schedulers: can't find dynamically referenced class androidx.work.impl.background.firebase.FirebaseJobScheduler
00:20:09.210 [QUIET] [system.out] Note: kotlin.internal.PlatformImplementationsKt: can't find dynamically referenced class kotlin.internal.jdk8.JDK8PlatformImplementations
00:20:09.210 [QUIET] [system.out] Note: kotlin.internal.PlatformImplementationsKt: can't find dynamically referenced class kotlin.internal.JRE8PlatformImplementations
00:20:09.210 [QUIET] [system.out] Note: kotlin.internal.PlatformImplementationsKt: can't find dynamically referenced class kotlin.internal.jdk7.JDK7PlatformImplementations
00:20:09.210 [QUIET] [system.out] Note: kotlin.internal.PlatformImplementationsKt: can't find dynamically referenced class kotlin.internal.JRE7PlatformImplementations
00:20:09.212 [QUIET] [system.out] Note: kotlin.jvm.internal.Reflection: can't find dynamically referenced class kotlin.reflect.jvm.internal.ReflectionFactoryImpl
00:20:09.228 [QUIET] [system.out] Note: okhttp3.internal.platform.AndroidPlatform: can't find dynamically referenced class com.android.org.conscrypt.SSLParametersImpl
00:20:09.228 [QUIET] [system.out] Note: okhttp3.internal.platform.AndroidPlatform: can't find dynamically referenced class org.apache.harmony.xnet.provider.jsse.SSLParametersImpl
00:20:09.228 [QUIET] [system.out] Note: okhttp3.internal.platform.AndroidPlatform$CloseGuard: can't find dynamically referenced class dalvik.system.CloseGuard
00:20:09.228 [QUIET] [system.out] Note: okhttp3.internal.platform.ConscryptPlatform: can't find dynamically referenced class org.conscrypt.ConscryptEngineSocket
00:20:09.229 [QUIET] [system.out] Note: okhttp3.internal.platform.Platform: can't find dynamically referenced class sun.security.ssl.SSLContextImpl
00:20:09.955 [QUIET] [system.out] Note: the configuration keeps the entry point 'androidx.core.graphics.drawable.IconCompatParcelizer { android.support.v4.graphics.drawable.IconCompat read(androidx.versionedparcelable.VersionedParcel); }', but not the descriptor class 'androidx.versionedparcelable.VersionedParcel'
00:20:09.955 [QUIET] [system.out] Note: the configuration keeps the entry point 'androidx.core.graphics.drawable.IconCompatParcelizer { void write(android.support.v4.graphics.drawable.IconCompat,androidx.versionedparcelable.VersionedParcel); }', but not the descriptor class 'androidx.versionedparcelable.VersionedParcel'

These aren't failures (unless your configuration has a problem).  As you can see, you have things listed from other libraries like okhttp as well.

Maybe, I missing the specific part in the output. I pasted a large part of the lower end here: http://pastebin.de/8421/

You need to add:

-dontwarn sun.misc.Unsafe

Thank you. This made the release build succeed.
Can you please explain which library (version) introduces the Unsafe class?

I am still facing this issue with WorkManager-1.0.0-beta01

2018-12-31 01:40:32.701 2947-2977/de.aurora.mggvertretungsplan.debug E/WM-WorkerFactory: Could not instantiate de.aurora.mggvertretungsplan.services.DownloadTimeTableWorker
    java.lang.IllegalAccessException: Class java.lang.Class<androidx.work.WorkerFactory> cannot access  method void de.aurora.mggvertretungsplan.services.DownloadTimeTableWorker.<init>(android.content.Context, androidx.work.WorkerParameters) of class java.lang.Class<de.aurora.mggvertretungsplan.services.DownloadTimeTableWorker>
        at java.lang.reflect.Constructor.newInstance0(Native Method)
        at java.lang.reflect.Constructor.newInstance(Constructor.java:343)
        at androidx.work.WorkerFactory.createWorkerWithDefaultFallback(WorkerFactory.java:92)
        at androidx.work.impl.WorkerWrapper.runWorker(WorkerWrapper.java:191)
        at androidx.work.impl.WorkerWrapper.run(WorkerWrapper.java:125)
        at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1167)
        at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:641)
        at java.lang.Thread.run(Thread.java:764)
2018-12-31 01:40:32.701 2947-2977/de.aurora.mggvertretungsplan.debug E/WM-WorkerWrapper: Could not create Worker de.aurora.mggvertretungsplan.services.DownloadTimeTableWorker

My code can be found at https://github.com/d-Rickyy-b/MGGVertretungsplan.

I'd be very grateful for any kind of support.

Okay... I feel pretty stupid now - it seems that my constructor was package-private and for workmanager to work, it needed to be public. Maybe the exception could point to that issue?

That is correct. Worker must have a public constructor unless you instantiate it via your own WorkerFactory.

@Aloy: Which version of WorkManager are you using? Can you reproduce this issue with alpha08 ?
Also are you changing the default WorkManager Configuration ? Can you please send us a code snippet ?

For me alpha07 solved the problem.

This just have happened to a user. We use a unique work to download assets, that can be more than 100. We usually don't have issue but this time the user got:
Fatal Exception: java.lang.IllegalStateException: Apps may not schedule more than 100 distinct jobs
       at android.os.Parcel.readException(Parcel.java:2013)
       at android.os.Parcel.readException(Parcel.java:1951)
       at android.app.job.IJobScheduler$Stub$Proxy.schedule(IJobScheduler.java:180)

Device Mi A1 with Android 8.1.0. WorkManager 1.0.0-alpha06

This is the code:
        val workers = assets.map { assetModel ->
            val data = Data.Builder()
                    .putString(ASSET_UUID, assetModel.uuid)
                    .putString(ASSET_URL, assetModel.url)
                    .putString(ASSET_VERSION, assetModel.version.toString())
                    .build()
            OneTimeWorkRequest.Builder(AssetDownloadWorker::class.java)
                    .setInputData(data)
                    .build()
        }

        val continuation = workManager?.beginUniqueWork(
                DOWNLOAD_WORK_NAME + mainUuid,
                ExistingWorkPolicy.KEEP,
                workers.toList())

#32: As noted earlier, this is fixed in newer versions of WorkManager.  Please update to the latest version.

Sorry, for the wrong report. Seems to work fine for me now.

We are using alpha08 and we are still getting this exception. We use the WorkManager to send events to our backend. So we can enque a lot of requests (more than 100) in a short period of time.

We don't know exactly how to reproduce this error but we see in crashlytics that two users are getting this issue lots of times. The date we have about this users and crashes:

User 1, 11 crashes: Android  7.1.1 Nubia NX591J
Fatal Exception: java.lang.IllegalStateException: Apps may not schedule more than 100 distinct jobs
       at android.os.Parcel.readException(Parcel.java:1692)
       at android.os.Parcel.readException(Parcel.java:1637)
       at android.app.job.IJobScheduler$Stub$Proxy.schedule(IJobScheduler.java:158)
       at android.app.JobSchedulerImpl.schedule(JobSchedulerImpl.java:42)
       at androidx.work.impl.background.systemjob.SystemJobScheduler.scheduleInternal(SystemJobScheduler.java:126)
       at androidx.work.impl.background.systemjob.SystemJobScheduler.schedule(SystemJobScheduler.java:95)
       at androidx.work.impl.Schedulers.schedule(Schedulers.java:99)
       at androidx.work.impl.utils.EnqueueRunnable.scheduleWorkInBackground(EnqueueRunnable.java:114)
       at androidx.work.impl.utils.EnqueueRunnable.run(EnqueueRunnable.java:86)
       at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1133)
       at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:607)
       at java.lang.Thread.run(Thread.java:761)

User 2, 17 crashes: Android 7.0 Asus X008D
Fatal Exception: java.lang.IllegalStateException: Apps may not schedule more than 100 distinct jobs
       at android.os.Parcel.readException(Parcel.java:1691)
       at android.os.Parcel.readException(Parcel.java:1636)
       at android.app.job.IJobScheduler$Stub$Proxy.schedule(IJobScheduler.java:158)
       at android.app.JobSchedulerImpl.schedule(JobSchedulerImpl.java:42)
       at androidx.work.impl.background.systemjob.SystemJobScheduler.scheduleInternal(SystemJobScheduler.java:126)
       at androidx.work.impl.background.systemjob.SystemJobScheduler.schedule(SystemJobScheduler.java:95)
       at androidx.work.impl.Schedulers.schedule(Schedulers.java:99)
       at androidx.work.impl.utils.EnqueueRunnable.scheduleWorkInBackground(EnqueueRunnable.java:114)
       at androidx.work.impl.utils.EnqueueRunnable.run(EnqueueRunnable.java:86)
       at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1133)
       at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:607)
       at java.lang.Thread.run(Thread.java:761)

Yeah.. a fix should be available in alpha09 shortly. The relevant issue is https://issuetracker.google.com/issues/115560696

This was happening on android.arch.work:work-runtime:1.0.0-alpha01

This is not still happening on version :1.0.0-alpha09

Hi, this still happening in 1.0.0-alpha09. Only devices with 7 and 8 versions of android are affected.

Fatal Exception: java.lang.IllegalStateException: Apps may not schedule more than 100 distinct jobs
       at android.os.Parcel.readException(Parcel.java:1967)
       at android.os.Parcel.readException(Parcel.java:1905)
       at android.app.job.IJobScheduler$Stub$Proxy.schedule(IJobScheduler.java:180)
       at android.app.JobSchedulerImpl.schedule(JobSchedulerImpl.java:44)
       at androidx.work.impl.background.systemjob.SystemJobScheduler.scheduleInternal(SystemJobScheduler.java:148)
       at androidx.work.impl.background.systemjob.SystemJobScheduler.schedule(SystemJobScheduler.java:117)
       at androidx.work.impl.Schedulers.schedule(Schedulers.java:99)
       at androidx.work.impl.utils.EnqueueRunnable.scheduleWorkInBackground(EnqueueRunnable.java:114)
       at androidx.work.impl.utils.EnqueueRunnable.run(EnqueueRunnable.java:86)
       at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1162)
       at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:636)
       at java.lang.Thread.run(Thread.java:764)

The code, that i used:

class App : MultiDexApplication() {
 
  override fun onCreate() {
    super.onCreate()
    Job.schedule()
  }
}

class Job(ctx: Context, params: WorkerParameters) : Worker(ctx, params) {

  override fun doWork(): Worker.Result {
    Log.d("-------------------Job is done!--------------------")
    return  Result.SUCCESS
  }

  companion object {
    private const val TAG = "my_job"

    fun schedule() {
      val workManager = WorkManager.getInstance()
      val work = PeriodicWorkRequest.Builder(Job::class.java, 1, DAYS).addTag(TAG).build()
      workManager.enqueueUniquePeriodicWork(TAG, KEEP, work)
    }
  }
}

In our case, seems that alpha-10 is not having this crash anymore. After four days in production no events have happened.

Ok I'm updated to alpha10, and watching for statistics now.

Updated to alpha10 and the crash persisted. Doesn't seem to be fixed, so reverting my app back to firebase jobdispatcher.

We had this problem on aplha08, aplha09 and now on alpha11. On alpha11 it is happening less than before, but there are still many crashes.

I'm removing WorkManager from our project and getting back to JobScheduler.

I can't believe that Google is not able to fix a base/core issue in this library for 11 versions straight, it is totally unusable with this bug.

Is there a way we can raise this ticket to someone more competent that is able to solve this problem???

Such a shame Google...

We have fixed all known issues around this. If you are still seeing this, feel free to send us a stacktrace and a sample app that reproduces the problem.

Sorry, I could not reproduce the issue myself. I lost a whole day trying to reproduce the issue and I could not do it, but it does not mean that the crash is not happening.

On 50K users using the app (with alpha11 library), we see that around 0.1% (50) of them have these crashes. It only happens on Android version 7 and 8.

We had a bigger crash rate before upgrading to alpha11, but the bug is still happening.

The only traces I have are these from Fabric:

Android 7
Fatal Exception: java.lang.IllegalStateException: Apps may not schedule more than 100 distinct jobs
       at android.os.Parcel.readException(Parcel.java:1701)
       at android.os.Parcel.readException(Parcel.java:1646)
       at android.app.job.IJobScheduler$Stub$Proxy.schedule(IJobScheduler.java:158)
       at android.app.JobSchedulerImpl.schedule(JobSchedulerImpl.java:42)
       at androidx.work.impl.background.systemjob.SystemJobScheduler.scheduleInternal(Unknown Source:148)
       at androidx.work.impl.background.systemjob.SystemJobScheduler.schedule(Unknown Source:117)
       at androidx.work.impl.Schedulers.schedule(Unknown Source:99)
       at androidx.work.impl.WorkManagerImpl.rescheduleEligibleWork(Unknown Source:550)
       at androidx.work.impl.utils.ForceStopRunnable.run(Unknown Source:74)
       at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1133)
       at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:607)
       at java.lang.Thread.run(Thread.java:762)
           
           
Android 7  
Fatal Exception: java.lang.IllegalStateException: Apps may not schedule more than 100 distinct jobs
       at android.os.Parcel.readException(Parcel.java:1691)
       at android.os.Parcel.readException(Parcel.java:1636)
       at android.app.job.IJobScheduler$Stub$Proxy.schedule(IJobScheduler.java:158)
       at android.app.JobSchedulerImpl.schedule(JobSchedulerImpl.java:42)
       at androidx.work.impl.background.systemjob.SystemJobScheduler.scheduleInternal(Unknown Source:148)
       at androidx.work.impl.background.systemjob.SystemJobScheduler.schedule(Unknown Source:117)
       at androidx.work.impl.Schedulers.schedule(Unknown Source:99)
       at androidx.work.impl.WorkManagerImpl.rescheduleEligibleWork(Unknown Source:550)
       at androidx.work.impl.utils.ForceStopRunnable.run(Unknown Source:74)
       at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1133)
       at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:607)
       at java.lang.Thread.run(Thread.java:761)
           
           
Android 8
Fatal Exception: java.lang.IllegalStateException: Apps may not schedule more than 100 distinct jobs
       at android.os.Parcel.readException(Parcel.java:1975)
       at android.os.Parcel.readException(Parcel.java:1913)
       at android.app.job.IJobScheduler$Stub$Proxy.schedule(IJobScheduler.java:180)
       at android.app.JobSchedulerImpl.schedule(JobSchedulerImpl.java:44)
       at androidx.work.impl.background.systemjob.SystemJobScheduler.scheduleInternal(Unknown Source:148)
       at androidx.work.impl.background.systemjob.SystemJobScheduler.schedule(Unknown Source:117)
       at androidx.work.impl.Schedulers.schedule(Unknown Source:99)
       at androidx.work.impl.WorkManagerImpl.rescheduleEligibleWork(Unknown Source:550)
       at androidx.work.impl.utils.ForceStopRunnable.run(Unknown Source:74)
       at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1162)
       at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:636)
       at java.lang.Thread.run(Thread.java:764)

           
It is happening on a wide range of devices (some of them):

Samsung Galaxy S7 (herolte),
Samsung Galaxy S7 (heroltebmc), Android 7.0
Samsung Galaxy Note5 (noblelte), Android 7.0
Samsung Galaxy S6 (zeroflte), Android 7.0
Samsung Galaxy S7 edge (hero2lte), Android 8.0
Motorola Moto G (5th Gen) (cedric), Android 8.1
Motorola Moto G(4) Plus (athene_f), Android 7.0
LGE V20 (elsa), Android 8.0
LGE LG K20 (lv517n), Android 7.0
dream2qltesq, Android 7.0
Huawei 荣耀畅玩 6X (HWBLN-H), Android 7.0
Sony Xperia Z5 (SO-01H), Android 7.0
Sony Xperia XZ Premium (SO-04J), Android 8.0

I think what might be happening is that some of the already scheduled jobs, may be stuck in a certain state.
You might want to push an update on the next version of your app, where you can cancelAllWork(), and reschedule all the Workers you need.

I have tried with cancelAllWorkByTag using the tag we used to create the jobs in the newest staged release but it did not help. We are still seeing the same crashes. We have also updated the lib to the latest (beta01) version.

cancelAllWorkByTag() only cancels work with tag.To be sure, you could cancelAllWork() [https://developer.android.com/reference/androidx/work/WorkManager.html#cancelAllWork()]
That way you are guaranteed to start from a clean slate, and then you can reschedule your work. Also, are you using a custom configuration when instantiating WorkManager ?

Yes, but we used a static constant in NotificationWorker.class to initialize it and used the same constant to cancel the work, so all workers should be stopped.

Here is the code:

        WorkManager.getInstance().cancelAllWorkByTag(NotificationWorker.TAG);

        PeriodicWorkRequest.Builder notificationsWorkBuilder = new PeriodicWorkRequest.Builder(NotificationWorker.class, NOTIFICATIONS_PERIOD, TimeUnit.MILLISECONDS);
        notificationsWorkBuilder.addTag(NotificationWorker.TAG);
        PeriodicWorkRequest notificationWork = notificationsWorkBuilder.build();
        WorkManager.getInstance().enqueueUniquePeriodicWork(NotificationWorker.TAG, ExistingPeriodicWorkPolicy.REPLACE, notificationWork);

If that is the only worker, then you are good. If there are other workers with other tags involved then using cancelAllWork() is what I would recommend.
Just to ensure that you are starting from a clean slate I would also try and call https://developer.android.com/reference/android/app/job/JobScheduler#cancelAll(), and reschedule work.

Yes, this is the only worker.  In the latest build I've added the cancelAllWork() call just to be sure that all work is cancelled (even we don't have more workers), but since we had a lot of crashes, I wasn't allowed to "play" anymore with WorkManager and we replaced it with JobScheduler for API23+ devices.

Hello,

We recently launched an update that includes "androidx.work:work-runtime:2.0.1", and we used it to schedule all our tasks now, in a satisfactory way. I just want to inform, (as you can see in the images attached below), that this problem keeps happening sporadically.

Keep in mind that the number of users affected in our case is small, since the app has several million active users, but it also starts to be worrisome, since the number of errors generated by each user is high.

The errors mainly occur in Android 6 and 7, and sometimes in 8, but especially in Android 7.

I do not expose implementation details, since the documentation of this SDK has been thoroughly studied, and in case there was a bad implementation, the number of errors would be infinitely greater.

Greetings.

We are having the same issue as in #52, (lots of crashes from small number of users) on WorkManager 1.0.1

Can you please add your notes to this bug report going forward.

https://issuetracker.google.com/issues/134058261

This bug report talks about this crash in the 1.x alpha era. So I just want to disambiguate the problem.

Updated workmanager from 1.0.0-beta02 to 1.0.1, as you recommended before, this should fix an issue.
Total crash count is raised drasticly, here we have stats for 40000 devices with 1.0.1 workmanager version: This issue has 46707 crashes affecting 347 users

Hi, unfortunately you're dealing with ~6 month old versions at this point.  You should upgrade to the latest version (2.2) - we won't be diagnosing issues on older versions anymore.

Hi, we're facing this issue too, also with v1.0.1.

We're using WorkManager#enqueue(OneTimeWorkRequest), instead of WorkManager#beginUniqueWork(OneTimeWorkRequest).

Upgrading to v2.x as you suggest forces us to migrate to Android X too. On v2.x I didn't see anything related in the changelog to justify the migration to Android X.. how can I be sure that migrating both Android X and work-runtime v2.x is gonna fix our issues?

Currently it's happening only in Android 7 (API 24), mainly in Motorola phones (70% Moto C Plus: https://www.gsmarena.com/motorola_moto_c_plus-8646.php)..

Thank you, regards!

We made a lot of improvements in 2.X apart from the AndroidX migrations. This problem should go away when you switch to WorkManager 2.2.0. We recommend updating to the latest version asap.

This is a bug.

the same problem continues to appear in androidx.work:work-runtime:2.0.1

androidx.work:work-runtime-ktx 2.5.0

Fatal Exception: java.lang.IllegalStateException
JobScheduler 100 job limit exceeded. We count 0 WorkManager jobs in JobScheduler; we have 1 tracked jobs in our DB; our Configuration limit is 20.
androidx.work.impl.background.systemjob.SystemJobScheduler.scheduleInternal (SystemJobScheduler.java:204)
androidx.work.impl.background.systemjob.SystemJobScheduler.schedule (SystemJobScheduler.java:132)
androidx.work.impl.Schedulers.schedule (Schedulers.java:108)
androidx.work.impl.utils.ForceStopRunnable.forceStopRunnable (ForceStopRunnable.java:176)
androidx.work.impl.utils.ForceStopRunnable.run (ForceStopRunnable.java:102)
androidx.work.impl.utils.SerialExecutor$Task.run (SerialExecutor.java:91)
java.util.concurrent.ThreadPoolExecutor.runWorker (ThreadPoolExecutor.java:1167)
java.util.concurrent.ThreadPoolExecutor$Worker.run (ThreadPoolExecutor.java:641)
java.lang.Thread.run (Thread.java:764)
Caused by java.lang.IllegalStateException
Apps may not schedule more than 100 distinct jobs
android.os.Parcel.createException (Parcel.java:1958)
android.os.Parcel.readException (Parcel.java:1918)
android.os.Parcel.readException (Parcel.java:1868)
android.app.job.IJobScheduler$Stub$Proxy.schedule (IJobScheduler.java:184)
android.app.JobSchedulerImpl.schedule (JobSchedulerImpl.java:44)
androidx.work.impl.background.systemjob.SystemJobScheduler.scheduleInternal (SystemJobScheduler.java:186)
androidx.work.impl.background.systemjob.SystemJobScheduler.schedule (SystemJobScheduler.java:132)
androidx.work.impl.Schedulers.schedule (Schedulers.java:108)
androidx.work.impl.utils.ForceStopRunnable.forceStopRunnable (ForceStopRunnable.java:176)
androidx.work.impl.utils.ForceStopRunnable.run (ForceStopRunnable.java:102)
androidx.work.impl.utils.SerialExecutor$Task.run (SerialExecutor.java:91)
java.util.concurrent.ThreadPoolExecutor.runWorker (ThreadPoolExecutor.java:1167)
java.util.concurrent.ThreadPoolExecutor$Worker.run (ThreadPoolExecutor.java:641)
java.lang.Thread.run (Thread.java:764)