One of my customer reported that vGPU enabled Windows 7 VM’s BSOD between 24 to 48hours. It runs ESXi6.0 on cisco UCSC-C240-M4SX hardware. ESXi has a nvidia driver version 352.70-10EM.600.0.0.2494585
Issue occur when he open up excel application. while running the dump crash noticed it crashes at nvlddmkm.sys
*** STOP: 0x00000050 (0xFFFFF8800C77C000,0x0000000000000001,0xFFFFF880040551BD,0x0000000000000000)
*** nvlddmkm.sys - Address FFFFF880040551BD base at FFFFF8800380A000, DateStamp5660ac24
Tried uninstall the Nvidia driver’s on the windows 7 machine and re-installed, still the issue occurs.
Old version of Nvidia driver also has a same problem. Customer confirmed that he has followed grid vgpu deployment guide to step up his environment.
need help to fix the issue or to find any hardware fault
What Desktop Agent is your customer running and what version is it ?
What Grid card are you using ?
What vGPU profile is your customer using ?
Do you see anything in the Event Log for the VM around the time it crashes ?
Are you able to capture a memory dump from the crash ?
Do you by chance have any power profiles set - if so make sure that the VM Power Settings are not set to Power Off the machine or the display.
You say that the problem happens after 24-48 hours but is triggered by use of Excel.
So do I understand that if you run Excel on the VM in the first 24 hours or so its OK and then between 24-48 hours running Excel causes the crash. ?
I thought I’d posted a note about the new drivers here, but it seems to not have gone live. They went live on Monday so thanks for testing so quickly!!