GPU Titan Xp Fan Error

Graphics / Linux Linux
1

Hi,

Our Titan XP ran into fan errors many times. After restarting the server, it still can work very well. Does anyone have insights on it?

Error message is here:
[url]http://hmshan.io/WechatIMG1362.png[/url]

Some system configures are

Ubuntu 16.04.3 LTS
CUDA version 9.0.176

Thanks.
nvidia-bug-report.log.gz (307 KB)

2

Can you check with a more current driver? If it still fails, please run nvidia-bug-report.sh as root and attach the resulting .gz file to your post. Hovering the mouse over an existing post of yours will reveal a paperclip icon.

3

Hi generix,

I have uploaded the nvidia-bug-report.log.gz file. Please help check it. Other lab members are using the current driver. If the problem comes with the driver, we can update it then.
nvidia-bug-report.log.gz (307 KB)

4

Looks like it starts when the driver hits an XID 62 and a display engine timeout. This might be due to the Titan is also used to run X while running high cuda loads. Try disabling the Xserver after reboot and check if the issue still appears.