EXC_BAD_ACCESS (SIGSEGV) in sentrycrashmc_getThreadAtIndex; SentryCrash Exception Handler (Primary)

[vorporeal]

Platform

macOS

Installed

Carthage

Version

7.20.0

Steps to Reproduce

We've had trouble consistently reproducing this issue. A user reported that our app (Warp, a MacOS terminal) crashes when they try to run complex Terraform commands; see warpdotdev/Warp#1619.

They provided a crash log (attached here, but also available from the linked issue) which indicates that the crashing thread is SentryCrash Exception Handler (Primary), with the following stack trace:

Thread 19 Crashed:: SentryCrash Exception Handler (Primary)
0   Sentry                        	       0x1095e33b8 sentrycrashmc_getThreadAtIndex + 4
1   Sentry                        	       0x109596830 writeAllThreads + 184
2   Sentry                        	       0x1095964a4 sentrycrashreport_writeStandardReport + 2088
3   Sentry                        	       0x1095b988c onCrash + 144
4   Sentry                        	       0x1095b6224 sentrycrashcm_handleException + 156
5   Sentry                        	       0x1095ba984 handleExceptions + 724
6   libsystem_pthread.dylib       	       0x18f45026c _pthread_start + 148
7   libsystem_pthread.dylib       	       0x18f44b08c thread_start + 8

We have seen this occur other times in the past (so it's not a 7.20.0 regression), but there hasn't ever been a consistent pattern, and we've never been able to reliably reproduce it.

While the implication here is that sentry is crashing during handling of a crash elsewhere in the stack (and so there would be user impact even with a fix in place), it would be great to fix this so that we get accurate reporting of the initial crash.

I'm not sure whether this would have any relevance, but our application is written primarily in Rust, with some Objective-C code, and we use both the Cocoa and Rust Sentry SDKs in our application simultaneously.

Expected Result

Our application doesn't crash with a stack trace that points to a segfault in Sentry code.

Actual Result

See attached crash log: crash-report.txt

[brustolin]

Thanks @vorporal. We are aware of this and working on a solution.

This is related to #1892

[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 🥀

[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 🥀

[brustolin]

This was fixed with #2939

[AbdulmalekAlshugaa]

Having the same issue aging with the new sentry update : SentryCrash Exception Handler (Secondary)

[AbdulmalekAlshugaa]

SentryCrash Exception Handler (Secondary)
0   libsystem_kernel.dylib        	       0x102158bb0 mach_msg2_trap + 8
1   libsystem_kernel.dylib        	       0x102169d44 mach_msg2_internal + 76
2   libsystem_kernel.dylib        	       0x102185504 thread_suspend + 108
3   HelloChefTest                 	       0x100a49c74 handleExceptions + 128
4   libsystem_pthread.dylib       	       0x10309b4c0 _pthread_start + 104
5   libsystem_pthread.dylib       	       0x1030966f0 thread_start + 8
Thread 14:: SentryCrash Exception Handler (Primary)
0   libsystem_kernel.dylib        	       0x102158bb0 mach_msg2_trap + 8
1   libsystem_kernel.dylib        	       0x102169d44 mach_msg2_internal + 76
2   libsystem_kernel.dylib        	       0x102160dd4 mach_msg_overwrite + 532
3   libsystem_kernel.dylib        	       0x102158f28 mach_msg + 20
4   HelloChefTest                 	       0x100a49ca0 handleExceptions + 172
5   libsystem_pthread.dylib       	       0x10309b4c0 _pthread_start + 104
6   libsystem_pthread.dylib       	       0x1030966f0 thread_start + 8

[philipphofmann]

@AbdulmalekAlshugaa, please open a new issue pointing to this issue and put in the required details of the bug template. This helps us to better understand what could cause this. Thanks.