HearthSim/HSTracker

HSTracker caused kernel panic?

Opened this issue · 6 comments

Describe the bug
I have had 2 crashes on HSTracker startup after a reboot with 2.6.0. Today's actually seems to have triggered a kernel panic.

To Reproduce
Steps to reproduce the behavior:
(unfortunately not reliably)

  1. Reboot
  2. Start HSTracker
  3. Start Battle.net, then Hearthstone
  4. Crash (sometimes Kernel Panic )

Game mode

  • Battlegrounds

Expected behavior
No Crashes

Screenshots
N/A

Version (please complete the following information):
2.6.0

Additional context
Add any other context about the problem here.

Log file
HStracker crash log and kernel panic log attached.

For the kernel panic:
kernel-crash.txt
hstracker-kernel-crash.txt

There is nothing I can do about the Kernel panic. That is an Apple issue, I suggest reporting it directly to Apple.

As for the HSTracker crash, it is somehow tied to loading HSTracker.app/Contents/Resources/Resources/Managed/mscorlib.dll. I am not sure why that would be the case, I expect the file to be present on your system. I will update to latest Mono for the next release, hopefully that helps.

There is nothing I can do about the Kernel panic. That is an Apple issue, I suggest reporting it directly to Apple.

I figured as much. I wonder if it's something do to with Game Mode.

As for the HSTracker crash, it is somehow tied to loading HSTracker.app/Contents/Resources/Resources/Managed/mscorlib.dll. I am not sure why that would be the case, I expect the file to be present on your system. I will update to latest Mono for the next release, hopefully that helps.

The file is present on my system. Thanks for checking!

This looks like the same issue I reported in #1330, and I just ran into another crash myself:

HSTracker-2023-11-14-145558.ips.txt

I don't know about the kernel panic though... weirdly enough it seems your kernel panic happened at 9:48:23 UTC while the HSTracker crash was recorded at 9:49:44 UTC, but the crash seems framebuffer-related, were you watching HDR video or using an external monitor?

I can't make sense of why Mono is failing to initialize:

Thread 2 Crashed:: tid_1103
0   ???                           	    0x7ff8a58fea78 ???
1   libsystem_kernel.dylib        	    0x7ff8154107ce __pthread_kill + 10
2   libsystem_pthread.dylib       	    0x7ff815448f30 pthread_kill + 262
3   libsystem_c.dylib             	    0x7ff815367a49 abort + 126
4   libcoreclr.dylib              	       0x110073fde mono_post_native_crash_handler + 14
5   libcoreclr.dylib              	       0x11001805a mono_handle_native_crash + 458
6   libcoreclr.dylib              	       0x110113818 sigabrt_signal_handler.cold.1 + 40
7   libcoreclr.dylib              	       0x1100736c8 sigabrt_signal_handler + 152
8   libsystem_platform.dylib      	    0x7ff81547637d _sigtramp + 29
9   ???                           	    0x2e6d65747379 ???
10  libsystem_c.dylib             	    0x7ff815367a49 abort + 126
11  libcoreclr.dylib              	       0x1100bc7b8 monoeg_assert_abort + 24
12  libcoreclr.dylib              	       0x10fefc7ea mono_log_write_logfile + 250
13  libcoreclr.dylib              	       0x1100bcd22 monoeg_g_logv_nofree + 194
14  libcoreclr.dylib              	       0x1100bce9f monoeg_assertion_message + 143
15  libcoreclr.dylib              	       0x1100bceda mono_assertion_message + 26
16  libcoreclr.dylib              	       0x10fe2cca7 mono_assembly_load_corlib + 567
17  libcoreclr.dylib              	       0x10fe26fc2 mono_init + 322
18  libcoreclr.dylib              	       0x10ff6f942 mini_init + 1282
19  libcoreclr.dylib              	       0x10ffd529e mono_jit_init + 14

This should fail everytime and not just occasionally.

If it helps, due to power outages I had a bunch of reboots today, and HSTracker crashed like this every time the first time I started it after a reboot, on two separate machines. It does not happen after that first startup.

That's so bizarre. The only thing I can think is that Mono has a bug but it hasn't changed in a while but I will update to the latest on the upcoming version and hope it improves things. This initialization code hasn't changed in a good long time.