Unity Application hangs if left alone for couple of hours

Unity Engine
, , , , ,
1

We are running a unity 3d application on a windows touchscreen computer with integrated graphics Intel (R) UHD Graphics 620. We noticed that when we leave the application alone for few hours and come back and touch it, it hangs and crashes.
There is nothing suspicious in the logs. In the Windows System Event Viewer, we noticed that the display driver igfx stopped responding and successfully recovered. This was few minutes before the application crash was recorded in the Windows App Event viewer.

image
image900×68 4.98 KB

This is the windows app log:

Level Date and Time Source Event ID Task Category
Information 7/23/2024 2:20:15 PM Microsoft-Windows-Winsrv 10002 None The following application was terminated because it was hung: ConfirMap.exe.
Information 7/23/2024 2:11:50 PM Windows Error Reporting 1001 None “Fault bucket , type 0
Event Name: LiveKernelEvent
Response: Not available
Cab Id: 0
Problem signature:
P1: 141
P2: ffffe6022f95b460
P3: fffff8023c436750
P4: 0
P5: 17a8
P6: 10_0_19044
P7: 0_0
P8: 256_1
P9:
P10:
Attached files:
\?\C:\Windows\LiveKernelReports\WATCHDOG\WATCHDOG-20240723-1411.dmp
\?\C:\Windows\TEMP\WER-13468765-0.sysdata.xml
\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER9F09.tmp.WERInternalMetadata.xml
\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER9F39.tmp.xml
\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER9F38.tmp.csv
\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER9F58.tmp.txt
These files may be available here:
\?\C:\ProgramData\Microsoft\Windows\WER\ReportArchive\Kernel_141_3f2a2d52ea06ca76955f95feb7cdfe45349a8_00000000_cab_19c88f71-dae0-4081-a106-a8e6b30bbcea
Analysis symbol:
Rechecking for solution: 0
Report Id: 19c88f71-dae0-4081-a106-a8e6b30bbcea
Report Status: 2049
Hashed bucket:
Cab Guid: 0”
Information 7/23/2024 2:11:50 PM Windows Error Reporting 1001 None “Fault bucket , type 0
Event Name: LiveKernelEvent
Response: Not available
Cab Id: 0
Problem signature:
P1: 141
P2: ffffe6022f95b460
P3: fffff8023c436750
P4: 0
P5: 17a8
P6: 10_0_19044
P7: 0_0
P8: 256_1
P9:
P10:
Attached files:
\?\C:\Windows\LiveKernelReports\WATCHDOG\WATCHDOG-20240723-1411.dmp
\?\C:\Windows\TEMP\WER-13468765-0.sysdata.xml
\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER9F09.tmp.WERInternalMetadata.xml
\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER9F39.tmp.xml
\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER9F38.tmp.csv
\?\C:\ProgramData\Microsoft\Windows\WER\Temp\WER9F58.tmp.txt
These files may be available here:
\?\C:\ProgramData\Microsoft\Windows\WER\ReportQueue\Kernel_141_3f2a2d52ea06ca76955f95feb7cdfe45349a8_00000000_19c88f71-dae0-4081-a106-a8e6b30bbcea
Analysis symbol:
Rechecking for solution: 0
Report Id: 19c88f71-dae0-4081-a106-a8e6b30bbcea
Report Status: 4
Hashed bucket:
Cab Guid: 0”

We updated the driver graphics and reduced the FPS of the application to no avail.

Any help or lead is appreciated.

2

Oh yeah, those integrated graphics can be a real nightmare! :frowning:

Updating to the latest driver may sometimes not be the best course of action but rather trying several versions, even very old ones as long as it doesn’t cause issues with your app.

There’s also a chance the CPU may be overheating and causing this reset. Especially if the system is in a closed box and/or the case itself is pretty small. Those typically “office line” machines aren’t designed to be running at full power for a prolongued time. It may help to underclock the system in BIOS for a test. Be sure to prefer all the “safe” settings over “performance”.

Definitely make sure you disable things like screensaver and display standby in Windows, and if that doesn’t help even disable all power-saving features in BIOS. Perhaps even dynamic clocking if that is possible, ie force the CPU to never go into “turbo” clocks but force it to stay on the base clock, neither going “turbo” nor “energy saving”.

I would also check on another system, ideally with the same integrated graphics whether the same issue occurs to see if this may be a graphics / CPU specific thing, or just a system that has some sort of instability.

3

Thank you for the prompt response. We see the same issue on other exact systems. It is a closed box Monitor with integrated CPU. To be specific Advantech POC-W213 21.5" Widescreen Point-of-Care Terminal with Fanless and Slim Design.
The temperature of the back panel reaches 135 degrees and Internal cpu temp is around 95-100. Will try out the settings you suggested. Will also try out on other systems to see any similar issue.

If you can think of any other things that can help, appreciate it.

4

You could also try lowering the quality settings and disable post fx, if only for testing. Perhaps one of these settings or effects ultimately causes the issue and once you find it, you could disable it. A first test would be to lower/disable everything to confirm that this does prevent crashes. And if so, you would go for a bug hunt, one setting at a time (if you’re unlucky it’s a combination of multiple settings though).

5

Just a quick update.

Testing it on multiple systems. No crashes to determine which fix works. Did the following for each of the system respectively - Reduce graphics settings, Update TDR, Used older Unity build 2022, etc.

In process of underclocking the CPU and update BIOS settings to see if it helps.