With 3.28.0e game is crashing ever few minutes

I'm trying a few more things. Another general update of my system, clearing shaders, verifying cache.

I did have one weird idea - I don't know if you all use KDE Plasma but I previously had "Switch to Dark Mode at Night" selected which I thought might be correlating with my "crashing more at night" idea. It was still in dark mode this morning too which supports it.

I turned myself back to "Breeze" for the morning theme. This wouldn't be the first time KDE Plasma screwed with stuff for me. It's a crazy idea but fingers crossed it helps!

EDIT: crashed after 1 hour. It lasted awhile but probably still just random. I installed GE-10.34 but I'm not hopeful.

EDIT2: here is the last 50 lines of my crash log. I don't think it helps, just seems like Wine noise.

Spoiler
91196.631:0118:0148:trace:unwind:dump_unwind_info unwind info at 00006FFFFF97800C flags 0 prolog 0x14 bytes function 00006FFFFF923300-00006FFFFF9239E0
91196.631:0118:0148:trace:unwind:dump_unwind_info frame register rbp offset 0x80(%rsp)
91196.631:0118:0148:trace:unwind:dump_unwind_info 0x14: leaq 0x80(%rsp),rbp
91196.631:0118:0148:trace:unwind:dump_unwind_info 0xc: subq $0xa80,%rsp
91196.631:0118:0374:trace:unwind:RtlGrowFunctionTable 0000000003A92300, 56
91196.631:0118:0148:trace:unwind:dump_unwind_info 0x5: pushq %r12
91196.631:0118:0148:trace:unwind:dump_unwind_info 0x3: pushq %r13
91196.631:0118:0148:trace:unwind:dump_unwind_info 0x1: pushq %rbp
91196.631:0118:0148:trace:unwind:RtlVirtualUnwind2 type 2 base 6fffff900000 rip 6fffff90529b rva 529b rsp 3d9fac0
91196.631:0118:0148:trace:unwind:dump_unwind_info **** func 5280-52a0
91196.631:0118:0148:trace:unwind:dump_unwind_info unwind info at 00006FFFFF977ED4 flags 0 prolog 0x4 bytes function 00006FFFFF905280-00006FFFFF9052A0
91196.631:0118:0148:trace:unwind:dump_unwind_info 0x4: subq $0x28,%rsp
91196.631:0118:0148:trace:unwind:RtlVirtualUnwind2 inside epilog.
91196.631:0118:0148:trace:unwind:RtlVirtualUnwind2 type 2 base 6fffffac0000 rip 6fffffac8b96 rva 8b96 rsp 3d9faf0
91196.631:0118:0148:trace:unwind:dump_unwind_info **** func 88a0-8e9f
91196.631:0118:0374:trace:unwind:RtlGrowFunctionTable 0000000003A92300, 57
91196.631:0118:0148:trace:unwind:dump_unwind_info unwind info at 00006FFFFFADA740 flags 0 prolog 0x17 bytes function 00006FFFFFAC88A0-00006FFFFFAC8E9F
91196.631:0118:0148:trace:unwind:dump_unwind_info frame register rbp offset 0x80(%rsp)
91196.631:0118:0148:trace:unwind:dump_unwind_info 0x17: leaq 0x80(%rsp),rbp
91196.631:0118:0148:trace:unwind:dump_unwind_info 0xf: subq $0x458,%rsp
91196.631:0118:0148:trace:unwind:dump_unwind_info 0x8: pushq %rbx
91196.631:0118:0148:trace:unwind:dump_unwind_info 0x7: pushq %rsi
91196.631:0118:0148:trace:unwind:dump_unwind_info 0x6: pushq %rdi
91196.631:0118:0148:trace:unwind:dump_unwind_info 0x5: pushq %r12
91196.631:0118:0148:trace:unwind:dump_unwind_info 0x3: pushq %r13
91196.631:0118:0148:trace:unwind:dump_unwind_info 0x1: pushq %rbp
91196.631:0118:0148:trace:seh:RtlRestoreContext returning to 00006FFFFFAC273A stack 0000000003D9FAF0
91196.631:0118:0374:trace:unwind:RtlGrowFunctionTable 0000000003A92300, 58
91196.631:0118:0148:trace:mscoree:DllMain (00006FFFFDE00000, 3, 0000000000000000)
91196.631:0118:0374:trace:unwind:RtlGrowFunctionTable 0000000003A92300, 59
91196.631:0030:0380:warn:threadname:NtSetInformationThread Thread renamed to L"wine_threadpool_worker"
91196.632:0118:011c:trace:unwind:RtlGrowFunctionTable 0000000003A92300, 60
91196.632:0118:011c:trace:unwind:RtlGrowFunctionTable 0000000003A92300, 61
91196.632:0118:011c:trace:unwind:RtlGrowFunctionTable 0000000003A92300, 62
91196.632:0030:0384:warn:threadname:NtSetInformationThread Thread renamed to L"wine_threadpool_worker"
91196.632:0030:0388:warn:threadname:NtSetInformationThread Thread renamed to L"wine_threadpool_worker"
91196.632:0118:037c:trace:mscoree:DllMain (00006FFFFDE00000, 3, 0000000000000000)
91196.633:0118:0374:trace:mscoree:DllMain (00006FFFFDE00000, 3, 0000000000000000)
91196.633:0118:0378:trace:mscoree:DllMain (00006FFFFDE00000, 3, 0000000000000000)
91196.633:0118:015c:trace:mscoree:DllMain (00006FFFFDE00000, 3, 0000000000000000)
91196.682:0118:011c:trace:mscoree:_CorDllMain (00000000049D0000, 0, 0000000000000001)
91196.682:0118:011c:trace:mscoree:_CorDllMain (0000000004550000, 0, 0000000000000001)
91196.682:0118:011c:trace:mscoree:_CorDllMain (0000000004500000, 0, 0000000000000001)
91196.682:0118:011c:trace:mscoree:_CorDllMain (0000000003C60000, 0, 0000000000000001)
91196.682:0118:011c:trace:mscoree:_CorDllMain (0000000003C40000, 0, 0000000000000001)
91196.682:0118:011c:trace:mscoree:_CorDllMain (0000000010000000, 0, 0000000000000001)
91196.682:0118:011c:trace:mscoree:_CorDllMain (0000000002E10000, 0, 0000000000000001)
91196.683:0118:011c:trace:mscoree:DllMain (00006FFFFDE00000, 0, 0000000000000001)
pid 261579 != 261578, skipping destruction (fork without exec?)
Last edited by Morganathon#4311 on Mar 27, 2026, 4:17:09 PM
"
I'm trying a few more things. Another general update of my system, clearing shaders, verifying cache.

I did have one weird idea - I don't know if you all use KDE Plasma but I previously had "Switch to Dark Mode at Night" selected which I thought might be correlating with my "crashing more at night" idea. It was still in dark mode this morning too which supports it.

I turned myself back to "Breeze" for the morning theme. This wouldn't be the first time KDE Plasma screwed with stuff for me. It's a crazy idea but fingers crossed it helps!

EDIT: crashed after 1 hour. It lasted awhile but probably still just random. I installed GE-10.34 but I'm not hopeful.

EDIT2: here is the last 50 lines of my crash log. I don't think it helps, just seems like Wine noise.

Spoiler
91196.631:0118:0148:trace:unwind:dump_unwind_info unwind info at 00006FFFFF97800C flags 0 prolog 0x14 bytes function 00006FFFFF923300-00006FFFFF9239E0
91196.631:0118:0148:trace:unwind:dump_unwind_info frame register rbp offset 0x80(%rsp)
91196.631:0118:0148:trace:unwind:dump_unwind_info 0x14: leaq 0x80(%rsp),rbp
91196.631:0118:0148:trace:unwind:dump_unwind_info 0xc: subq $0xa80,%rsp
91196.631:0118:0374:trace:unwind:RtlGrowFunctionTable 0000000003A92300, 56
91196.631:0118:0148:trace:unwind:dump_unwind_info 0x5: pushq %r12
91196.631:0118:0148:trace:unwind:dump_unwind_info 0x3: pushq %r13
91196.631:0118:0148:trace:unwind:dump_unwind_info 0x1: pushq %rbp
91196.631:0118:0148:trace:unwind:RtlVirtualUnwind2 type 2 base 6fffff900000 rip 6fffff90529b rva 529b rsp 3d9fac0
91196.631:0118:0148:trace:unwind:dump_unwind_info **** func 5280-52a0
91196.631:0118:0148:trace:unwind:dump_unwind_info unwind info at 00006FFFFF977ED4 flags 0 prolog 0x4 bytes function 00006FFFFF905280-00006FFFFF9052A0
91196.631:0118:0148:trace:unwind:dump_unwind_info 0x4: subq $0x28,%rsp
91196.631:0118:0148:trace:unwind:RtlVirtualUnwind2 inside epilog.
91196.631:0118:0148:trace:unwind:RtlVirtualUnwind2 type 2 base 6fffffac0000 rip 6fffffac8b96 rva 8b96 rsp 3d9faf0
91196.631:0118:0148:trace:unwind:dump_unwind_info **** func 88a0-8e9f
91196.631:0118:0374:trace:unwind:RtlGrowFunctionTable 0000000003A92300, 57
91196.631:0118:0148:trace:unwind:dump_unwind_info unwind info at 00006FFFFFADA740 flags 0 prolog 0x17 bytes function 00006FFFFFAC88A0-00006FFFFFAC8E9F
91196.631:0118:0148:trace:unwind:dump_unwind_info frame register rbp offset 0x80(%rsp)
91196.631:0118:0148:trace:unwind:dump_unwind_info 0x17: leaq 0x80(%rsp),rbp
91196.631:0118:0148:trace:unwind:dump_unwind_info 0xf: subq $0x458,%rsp
91196.631:0118:0148:trace:unwind:dump_unwind_info 0x8: pushq %rbx
91196.631:0118:0148:trace:unwind:dump_unwind_info 0x7: pushq %rsi
91196.631:0118:0148:trace:unwind:dump_unwind_info 0x6: pushq %rdi
91196.631:0118:0148:trace:unwind:dump_unwind_info 0x5: pushq %r12
91196.631:0118:0148:trace:unwind:dump_unwind_info 0x3: pushq %r13
91196.631:0118:0148:trace:unwind:dump_unwind_info 0x1: pushq %rbp
91196.631:0118:0148:trace:seh:RtlRestoreContext returning to 00006FFFFFAC273A stack 0000000003D9FAF0
91196.631:0118:0374:trace:unwind:RtlGrowFunctionTable 0000000003A92300, 58
91196.631:0118:0148:trace:mscoree:DllMain (00006FFFFDE00000, 3, 0000000000000000)
91196.631:0118:0374:trace:unwind:RtlGrowFunctionTable 0000000003A92300, 59
91196.631:0030:0380:warn:threadname:NtSetInformationThread Thread renamed to L"wine_threadpool_worker"
91196.632:0118:011c:trace:unwind:RtlGrowFunctionTable 0000000003A92300, 60
91196.632:0118:011c:trace:unwind:RtlGrowFunctionTable 0000000003A92300, 61
91196.632:0118:011c:trace:unwind:RtlGrowFunctionTable 0000000003A92300, 62
91196.632:0030:0384:warn:threadname:NtSetInformationThread Thread renamed to L"wine_threadpool_worker"
91196.632:0030:0388:warn:threadname:NtSetInformationThread Thread renamed to L"wine_threadpool_worker"
91196.632:0118:037c:trace:mscoree:DllMain (00006FFFFDE00000, 3, 0000000000000000)
91196.633:0118:0374:trace:mscoree:DllMain (00006FFFFDE00000, 3, 0000000000000000)
91196.633:0118:0378:trace:mscoree:DllMain (00006FFFFDE00000, 3, 0000000000000000)
91196.633:0118:015c:trace:mscoree:DllMain (00006FFFFDE00000, 3, 0000000000000000)
91196.682:0118:011c:trace:mscoree:_CorDllMain (00000000049D0000, 0, 0000000000000001)
91196.682:0118:011c:trace:mscoree:_CorDllMain (0000000004550000, 0, 0000000000000001)
91196.682:0118:011c:trace:mscoree:_CorDllMain (0000000004500000, 0, 0000000000000001)
91196.682:0118:011c:trace:mscoree:_CorDllMain (0000000003C60000, 0, 0000000000000001)
91196.682:0118:011c:trace:mscoree:_CorDllMain (0000000003C40000, 0, 0000000000000001)
91196.682:0118:011c:trace:mscoree:_CorDllMain (0000000010000000, 0, 0000000000000001)
91196.682:0118:011c:trace:mscoree:_CorDllMain (0000000002E10000, 0, 0000000000000001)
91196.683:0118:011c:trace:mscoree:DllMain (00006FFFFDE00000, 0, 0000000000000001)
pid 261579 != 261578, skipping destruction (fork without exec?)

It cannot be this night mode, because I'm not using it on KDE.

I've tried PROTON_WAYLAND_ENABLE=1, with different runners and different renderers (DX11/DX12/VK). Still crashes.

What helps for me is:
1. go to steamapps/common/Path of Exile -> delete ShaderCacheD3D11/12/VK
2. go to steam -> right click -> files -> verify integrity
3. wait for it to finish
4. click play and it should work atleast for 1-2h
Also crashing, switched from Mint to CachyOS to see that helped and it worked for a bit but after a few hours started crashing again :/
Just adding my system specs here as I've been experiencing the same random crashes on 3.280e and before that patch. Really can't pinpoint anything that made it start to happen.

I'm on Linux, Bazzite, Fedora 43, kernel 6.17.7
NVIDIA 4080 Super - NVIDIA driver version: 595.45.04
Ryzen 9800 X3D
32GB of $$RAM$$

I play on the stand alone client through Heroic Games Launcher.

I've posted in both of these threads along with other that are experiencing the same issue on Linux and also on Windows.

https://www.pathofexile.com/forum/view-thread/3924743
https://www.pathofexile.com/forum/view-thread/3917556
Thanks for your input.

I'm still crashing like crazy. Nothing helps really.
I am on bazzite linux and am crashing every few minutes as well. It crashes in hideout, on maps, everywhere. No logs or anything just crashes to desktop. I crashed in proton 10, 9, experimental and hotfix. I crashed in vulkan and dx12. I even changed to fsr and still crashed. I hope we hear something soon I would like to finish these challenges.

Oh also I have 7900xtx gpu, 7800x3d cpu, 32 gigs of ram if that helps.
Last edited by Soulknotica#5136 on Apr 1, 2026, 7:41:10 AM
"
I am on bazzite linux and am crashing every few minutes as well. It crashes in hideout, on maps, everywhere. No logs or anything just crashes to desktop. I crashed in proton 10, 9, experimental and hotfix. I crashed in vulkan and dx12. I even changed to fsr and still crashed. I hope we hear something soon I would like to finish these challenges.

Oh also I have 7900xtx gpu, 7800x3d cpu, 32 gigs of ram if that helps.


Exactly the same experience. In-game settings doesn't matter, proton version doesn't matter.

Client.txt doesn't show anything, protonlog doesn't have anything.

Just crash to desktop.

EDIT:
I had enough today. I booted up my win11 disk and played there (DX12, all default ingame settings). It didn't crashed once. I've played for 2h maybe.
Last edited by TaresPL#2094 on Apr 1, 2026, 12:05:10 PM
Guys!

I got something! I'm playing for more than 24h without a single crash.

I've uninstalled poe completely. I've deleted manually every "Path of Exile" and 238960 folder I could find.

After that normal install, I've reconfigured everything in game manually (keybinds and game options), changed back to cachyos-proton-slr and that's it! (graphical settings are all default with FPS lock to 100).

I don't know why it helped, but it did. I can even enable MTX now, before it was not possible as I would crash twice as often.
NixOS (unstable channel) here, also crashing incredibly often, completely at random, no possible pattern that I could identify.

Only thing is that crashes seem to come in "crises", what I mean is I can play for a long time (2 or 3 hours), but the second I crash I will not be able to go on without crashing every 1 or 2 minutes for quite some time until it somehow becomes stable again.

EDIT: enabled PROTON_LOG=1 and, as others have pointed out, this is also my last line:

"
pid 71401 != 71400, skipping destruction (fork without exec?)
Last edited by SmokeMax#2893 on Apr 3, 2026, 10:02:26 PM
"
Pehr#3739 wrote:
I don't mean to detract from the problem you are experiencing, but for what it's worth Patch 3.28.0e hasn't been released yet.


'e' most certainly HAS been released.
Every time I launch the game it 'patches' to version 'e' again.

And again

And again

And each time I log in it fails to join instance. None of the conventional workarounds are working (i.e. change video mode, change servers, ect).


SOMEONE borked the absolute SNOT out of their patch. Hopefully someone will give up their holiday weekend to come revert everything back to version 'd' before the userbase has a collective meltdown.
Patch Notes 3.15:
Fixed a bug where players believed the game was playable. This has been corrected and made retroactive.
Patch Notes 3.19:
Fixed a bug where players adapted to 3.15. This bug cannot be corrected, so we have implemented a 90% reduction in item access as a punishment.

Report Forum Post

Report Account:

Report Type

Additional Info