Crash in monitorCachedData

[stefanfuerst]

Platform

macOS

Installed

Manually

Version

7.8

Steps to Reproduce

Lately I'm often getting C++ Exception: _NSCoreDataException that look like this

libsystem_kernel    0x1bdbf4ebc  __semwait_signal
libsystem_c         0x1bdaffd84  nanosleep
libsystem_c         0x1bdb0a6e8  sleep
Sentry              0x1010a9384  monitorCachedData (Sources/SentryCrash/Recording/SentryCrashCachedData.c:139)
libsystem_pthread   0x1bdc2d23c  _pthread_start

I can't tell if this is not parsed properly or an issue with Sentry. Also I was not able to reproduce this locally.

Expected Result

Display of details of the _NSCoreDataException

Actual Result

Just C++ Exception: _NSCoreDataException

[stefanfuerst]

I did some experiments by creating an exception on my own

This ends up in Sentry as C++ Exception: _NSCoreDataException without stack trace

- (void)applicationDidFinishLaunching:(NSNotification *)aNotification
{
    dispatch_after(dispatch_time(DISPATCH_TIME_NOW, (int64_t)(4 * NSEC_PER_SEC)), dispatch_get_global_queue(DISPATCH_QUEUE_PRIORITY_DEFAULT, 0), ^{
        [[NSClassFromString(@"_NSCoreDataException") exceptionWithName:@"Oh Damn" reason:@"Crash to crash" userInfo:nil] raise];
    });
}

While generating a regular Exception is reported with the correct parameters and stack trace

- (void)applicationDidFinishLaunching:(NSNotification *)aNotification
{
    dispatch_after(dispatch_time(DISPATCH_TIME_NOW, (int64_t)(4 * NSEC_PER_SEC)), dispatch_get_global_queue(DISPATCH_QUEUE_PRIORITY_DEFAULT, 0), ^{
        [NSException exceptionWithName:@"Oh Damn" reason:@"Crash to crash" userInfo:nil] raise];
    });
}

I did not yet figure out the point where reporting goes wrong

[brustolin]

Hey @stefanfuerst, thanks for letting us know this.
I will take a look.

[philipphofmann]

@brustolin was able to reproduce this. We can't give you an ETA for fixing this, though @stefanfuerst.

[github-actions]

This issue has gone three weeks without activity. In another week, I will close it.

But! If you comment or otherwise update it, I will reset the clock, and if you label it Status: Backlog or Status: In Progress, I will leave it alone ... forever!

---

"A weed is but an unloved flower." ― Ella Wheeler Wilcox 🥀

[philipphofmann]

This could be related to #1533 and might be fixed with #1533 (comment)

[philipphofmann]

We see rare occurrences of this in our internal SDK crashes.

[grzegorzkrukowski]

Is there any update on this one ? it keeps happening in our Application on the latest version still

[brustolin]

Sorry @grzegorzkrukowski, no updates on this one yet.

[rickarddahlvolvocars]

Any updates on this?

We get this crash rarely, so far only for one user, 5 crashes.
It happens on iOS 16.6.1
We user Sentry SPM with version 8.38.

          Thread
0  libsystem_kernel.dylib         0x14ac __semwait_signal + 8
1  libsystem_c.dylib              0x541c nanosleep + 216
2  libsystem_c.dylib              0x16058 sleep + 48
3  **<OUR APP NAME>**                     0x6dcae88 monitorCachedData + 660
4  libsystem_pthread.dylib        0x30ec _pthread_start + 116
5  libsystem_pthread.dylib        0x172c thread_start + 8

and

          SentryCrash Exception Handler (Secondary)
0  libsystem_kernel.dylib         0x1030 mach_msg2_trap + 8
1  libsystem_kernel.dylib         0x12b18 mach_msg2_internal + 76
2  libsystem_kernel.dylib         0x12db8 mach_msg_overwrite + 484
3  libsystem_kernel.dylib         0x1524 mach_msg + 20
4  **<OUR APP NAME>**                     0x6db137c handleExceptions + 172
5  libsystem_pthread.dylib        0x30ec _pthread_start + 116
6  libsystem_pthread.dylib        0x172c thread_start + 8
        

[philipphofmann]

@rickarddahlvolvocars, I'm sorry, but we haven't gotten to this yet. Thanks for the details.