React Native Android thread crashes in production (java.util.concurrent.RejectedExecutionException)

[davidshariff]

Description
When running FFmpegKit.execute or FFmpegKit.executeWithArguments on React Native Android, the app crashes abruptly in com.arthenica.ffmpegkit.reactnative.FFmpegKitReactNativeModule.ffmpegSessionExecute throwing java.util.concurrent.RejectedExecutionException.

Expected behavior
Should resolve with the FFMPEG output or throw an error.

Current behavior
Crashes the app.

To Reproduce
I'm currently running the following command:

FFmpegKit.executeWithArguments([
    '-i', uri,
    '-ac', '1',
    '-vn',
    '-c:a', 'flac',
    outputFilePath
])

Screenshots
Devices having issues:

Logs

java.util.concurrent.RejectedExecutionException: 
  at java.util.concurrent.ThreadPoolExecutor$AbortPolicy.rejectedExecution (ThreadPoolExecutor.java:2086)
  at java.util.concurrent.ThreadPoolExecutor.reject (ThreadPoolExecutor.java:848)
  at java.util.concurrent.ThreadPoolExecutor.execute (ThreadPoolExecutor.java:1394)
  at java.util.concurrent.AbstractExecutorService.submit (AbstractExecutorService.java:118)
  at com.arthenica.ffmpegkit.reactnative.FFmpegKitReactNativeModule.ffmpegSessionExecute (FFmpegKitReactNativeModule.java:595)
  at java.lang.reflect.Method.invoke (Native Method)
  at com.facebook.react.bridge.JavaMethodWrapper.invoke (JavaMethodWrapper.java:372)
  at com.facebook.react.bridge.JavaModuleWrapper.invoke (JavaModuleWrapper.java:151)
  at com.facebook.react.bridge.queue.NativeRunnable.run (Native Method)
  at android.os.Handler.handleCallback (Handler.java:938)
  at android.os.Handler.dispatchMessage (Handler.java:99)
  at com.facebook.react.bridge.queue.MessageQueueThreadHandler.dispatchMessage (MessageQueueThreadHandler.java:27)
  at android.os.Looper.loop (Looper.java:246)
  at com.facebook.react.bridge.queue.MessageQueueThreadImpl$4.run (MessageQueueThreadImpl.java:226)
  at java.lang.Thread.run (Thread.java:923)

Environment

• Platform: ReactNative Android
• Architecture: arm64-v8a, armeabi-v7a, armeabi
• Version: v4.5.1.full-gpl-lts
• Source branch: main
• react-native info: React: 16.13.1, ffmpeg-kit-react-native: 4.5.1

[tanersener]

Is this issue reproducible? I don't see a clear explanation in the stack trace about why the execution is rejected.

[mars-lan]

Additional info in my case:

Fatal Exception: java.util.concurrent.RejectedExecutionException
Task java.util.concurrent.FutureTask@72f1039 rejected from java.util.concurrent.ThreadPoolExecutor@b11767e[Terminated, pool size = 0, active threads = 0, queued tasks = 0, completed tasks = 0]

Seems like the executor has been shutdown already? https://docs.oracle.com/javase/7/docs/api/java/util/concurrent/ThreadPoolExecutor.html

[tanersener]

@mars-lan Thanks. Is this behaviour reproducible?

[mars-lan]

@tanersener unfortunately it's kind of random and not readily reproducible. However, it's frequent enough that it's one of the top crashing issues on our Crashlytics.

[tanersener]

We don't see it in the test apps. So, hard to make a comment about it.

[mars-lan]

I found the root cause and was able to reproduce it.

This line shuts down the executor when the main activity is destroyed. In many cases, the application process will also be killed so the cleanup is justified. However, if the application has any other services running, e.g. a foreground service, the process won't be killed. As a result, the executor won't be recreated when the main activity is launched again, thus leading to the submission of tasks to a terminated executor.

[tanersener]

Thanks for the analysis. Let me check the lifecycle documentation to verify your theory.

[tanersener]

I removed that line in the new v4.5.2 release we have for react-native. It should be fixed now.

[davidshariff]

I updated v4.5.2 to production, looks like the bug is confirmed fixed.

@tanersener I think we can resolve this issue now. Thank you.

[tanersener]

@davidshariff Great, thanks!