Nvidia OPEN GL crash on several systems, client flakes out
Answered
Hello,
If you have any suggestions or know of a bug/fix/issue, please share.
DW version 4.2.0.32842
I noticed some Nvidia and OPEN GL related errors on a few systems we deployed several months back.
- an error code 3 box pops up and the client turns from black to white and the window says not responding.
-
event 4101Display driver nvlddmkm stopped responding and has successfully recovered.The description for Event ID 1 from source NVIDIA OpenGL Driver cannot be found. Either the component that raises this event is not installed on your local computer or the installation is corrupted. You can install or repair the component on the local computer.
If the event originated on another computer, the display information had to be saved with the event.
The following information was included with the event:
The NVIDIA OpenGL driver detected a problem with the display
driver and is unable to continue. The application must close.
Event log at link attached showing 2 days of admin issues
https://ovsusa.sharefile.com/d-s9bffc03203f244318813faf2fee9a6b0
Error code: 3
Visit http://www.nvidia.com/page/support.html for more information.
The message resource is present but the message was not found in the message table
-
Doing some more legwork on this.. the systems with the issue are Nvidia Geforce GT210 based cards. I'm sorting that they are dated and not well supported anymore.
We will be flipping to a 730 based card.
Just found it odd that this appeared since about January 2022 as far as I can determine.
-
Hi Matt Barber,
Typically, the Nvidia GeForce GT210 should work, since, according to THIS site, they use OpenGL 3.3, and we require OpenGL 2.1 or newer.
First thing I would check is if there are newer drivers available then the one that is used. Preferbly from the original manufacturer, not the default ones that Windows uses.
Regarding the error codes.
I checked for the error 3 code, and found the following:
Solution for Nvidia GeForce OpenGL Error code 3
OpenGL Error Code 3 is generally related to overheating. Nvidia GeForce 210 is factory tweaked to offer you superior gaming experience. Any attempt to over-clock the card will cause heating issues and generate errors. The solution to this problem is tweaking settings closer to the factory settings.
Given the age of the GPU card, a good clean of the fan and perhaps some new or better thermal paste might do the job (for a while). I'm not familiar with these GPU so can't judge what and if you can tweak some settings or even factory default the device in case someone modified the settings.
Regarding the event 4101, this seems a recurring issue according to THIS link. Also, a possible solution is mentioned. You can try these, but at your own risk. I can't judge the impact of the recommended fixes, but it seems fairly save.
In any case we do recommend to always set up a restore point before executing such modifications.
Regarding January 2022, just a minor update could break (driver) compatibility, especially with older hardware components.
-
Just a follow up on this to provide info.
I have followed the suggestions of modifying the video card IRQ details
Help provided in the link you shared. https://www.flofaber.com/windows-nvlddmk-error-4101
Tell windows to use Message-Signaled Based Interrupts for your GPU.
It took a learning curve to add it all up.. but basically find the device instance ID of the video card, use that info to drill into the registry and add the MSI info for the card there and turn it on.
Look in the device manager IRQ listing to find the card and note if it has a positive # on it. Once you reboot go here again and see if the number has become a negative number.
I have done this on 15 systems or so. It seems to help. I have had one system have the issue again, but it pushed the failure from every few days to a couple weeks. I also set a scheduled reboot on that pc to help avoid the failure. (replacing the video card is a fallback in this case)
Note.. the device instance ID was the same on most systems, but was not identical on all. So I had to modify the registry entries to match the device instance ID on those systems.
Some details below if you want for reference.
Device ID:
PCI\VEN_10DE&DEV_0A65&SUBSYS_13123842&REV_A2\4&3507A271&0&0008
Registry info:
Computer\HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Enum\PCI\VEN_10DE&DEV_0A65&SUBSYS_13123842&REV_A2\4&3507a271&0&0008\Device Parameters\Interrupt Management\MessageSignaledInterruptProperties
Add this value and change to 1 then reboot.
MSISupported 1
Please sign in to leave a comment.
Comments
3 comments