BUG: NullPointerException in CallClient$contextClient$1.b

[Bruce-7-tech]

Our app experienced a crash due to a NullPointerException in the method co.daily.CallClient$contextClient$1.b. This appears to originate from within the Daily SDK, likely during an internal callback or context access.
the stacktraces are:
Fatal Exception: java.lang.NullPointerException
co.daily.CallClient$contextClient$1.b (Unknown Source:32)
android.os.Handler.handleCallback (Handler.java:995)
android.os.Handler.dispatchMessage (Handler.java:105)
android.os.Looper.loopOnce (Looper.java:288)
android.os.Looper.loop (Looper.java:393)
android.app.ActivityThread.main (ActivityThread.java:9535)
java.lang.reflect.Method.invoke (Method.java)
com.android.internal.os.RuntimeInit$MethodAndArgsCaller.run (RuntimeInit.java:600)
com.android.internal.os.ZygoteInit.main (ZygoteInit.java:1005)

[marcus-daily]

Hi @Bruce-7-tech, which version of the SDK are you using? Can you give more details about what was happening at the time of the crash, and also any messages which were logged around when the crash happened? Thank you!

[Bruce-7-tech]

The version is 0.30.0. the whole is "co.daily:client:0.30.0".
The scenario is as follows:
User A and User B are on an ongoing call. During the call, User A repeatedly turns the camera on and off—not switching between front and rear cameras, but enabling and disabling the camera itself in rapid succession.
After a short while, the app crashes.
The crash stack trace matches the bug report (a full stack trace captured at the moment of the crash is also attached).
You can see the stack traces from the attached file from firebase Crashlytics
Thank you!

crash-stacktraces.txt

[Bruce-7-tech]

Hi, just wondering if there’s any update on this issue?
I’d really appreciate any feedback or guidance. Thank you!

[marcus-daily]

@Bruce-7-tech I haven't been able to reproduce the issue here unfortunately. The attached stack trace implies that release() is being called on the CallClient during the call -- can you whether you're using this please? (or, if you're passing a Lifecycle into the constructor?)

Also, please could you let me know what kind of device you're testing on, and which Android version?

[Bruce-7-tech]

Hi Marcus-daily,

We’ve observed that this crash is occurring quite frequently in our production app, and we'd like to better understand the root cause and explore ways to mitigate it.

We're using CallClient with only the Application context passed in. We are not using any Lifecycle when initializing or managing the client.

We are already calling CallClient.release() inside the ViewModel.onCleared() method, so under normal circumstances, resources should be properly cleaned up. However, this crash still happens on various devices, including Android 15 (e.g., iQOO Neo10), as reported in Firebase.

Given that we’re following the recommended cleanup procedure, we’d appreciate your insights on:

What is the actual root cause of this crash?

Is there any underlying issue in the SDK that could lead to this, even when release() is called correctly?

Could the SDK handle this situation more gracefully to prevent the app from crashing?

This issue is impacting our users, so we’d really appreciate a timely resolution or workaround. Let me know if you need more logs or device details.

Thanks in advance!

[marcus-daily]

@Bruce-7-tech Unfortunately the root cause isn't clear from the stack trace -- it looks like it's a timing issue involving the release() method. We haven't encountered this issue in our automated testing and haven't received any similar reports.

If the cause of this is rapidly toggling the camera, I suspect there may be an underlying system/camera issue on some devices which we're failing to recover from.

I can put a potential fix in the next release of daily-android, but either way I'd recommend refactoring your app to create a single CallClient which never gets released. A CallClient can be used for multiple calls, and should consume minimal resources when no call is active (and the mic and camera inputs are disabled).

[Bruce-7-tech]

Thank you very much for the detailed response and for looking into this issue.

We appreciate your insights, and the explanation makes sense regarding potential system or camera-related issues on certain devices.

Could you let us know when the potential fix might be included in an upcoming release of daily-android? We’d love to track it and verify the behavior once it's available.

Thanks again for your support and helpful guidance!

[Bruce-7-tech]

@marcus-daily I am looking forward to the new version and would like to have a try.

[marcus-daily]

@Bruce-7-tech I imagine the next version will be available in the next couple of weeks. In the meantime I'd recommend applying the CallClient workaround to your app -- please let me know if this works!

[Bruce-7-tech]

@Bruce-7-tech I imagine the next version will be available in the next couple of weeks. In the meantime I'd recommend applying the CallClient workaround to your app -- please let me know if this works!

I’ll try refactoring the usage of CallClient in the project to a global singleton, and then verify whether the previous crash issue still occurs. I’ll update here with the latest progress afterward. Thanks!

[marcus-daily]

Thanks @Bruce-7-tech. Just to let you know, version 0.32.0 was released at the end of last week with a potential fix for this issue.

[Bruce-7-tech]

Thanks @Bruce-7-tech. Just to let you know, version 0.32.0 was released at the end of last week with a potential fix for this issue.

That sounds great! I will have a try.