ERROR: nvidia-settings could not find the registry key file

Accelerated Computing CUDA CUDA Setup and Installation
1

Dear experts,

for my Tesla M2070 on CentOS 7, 64bit, I installed nvidia driver according to http://www.nvidia.com/download/driverResults.aspx/118959/en-us.

However, upon launching nvidia-settings (through ssh -X tunnel) I got this error below. Any help what could be wrong ?

milias@adf1:~/.nvidia-settings 

** (nvidia-settings:4799): WARNING **: Couldn't connect to accessibility bus: Failed to connect to socket /tmp/dbus-YXp6oepPsC: Connection refused

ERROR: nvidia-settings could not find the registry key file. This file should have been installed along with this driver at either
       /usr/share/nvidia/nvidia-application-profiles-340.102-key-documentation or /usr/share/nvidia/nvidia-application-profiles-key-documentation. The application profiles
       will continue to work, but values cannot be prepopulated or validated, and will not be listed in the help text. Please see the README for possible values and
       descriptions.

milias@adf1:~/.nvidia-smi              
Sun May 21 00:24:49 2017       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 375.66                 Driver Version: 375.66                    |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  Tesla M2070         Off  | 0000:1A:00.0     Off |                    0 |
| N/A   N/A    P0    N/A /  N/A |      0MiB /  5301MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID  Type  Process name                               Usage      |
|=============================================================================|
|  No running processes found                                                 |
+-----------------------------------------------------------------------------+
milias@adf1:~/.nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2016 NVIDIA Corporation
Built on Tue_Jan_10_13:22:03_CST_2017
Cuda compilation tools, release 8.0, V8.0.61
milias@adf1:~/.
2

What looks like a reasonable reason:

3

You can resolve this issue by performing the following:

  1. See what version of the nvidia application profiles were installed
$ cd /usr/share/nvidia
$ ls -la  
total 36
drwxr-xr-x   2 root root  4096 Aug 23 07:28 .
drwxr-xr-x 283 root root 12288 Sep  5 18:11 ..
-rw-r--r--   1 root root  7110 Jul 12 03:05 nvidia-application-profiles-470.57.02-key-documentation
-rw-r--r--   1 root root  9082 Jul 12 03:05 nvidia-application-profiles-470.57.02-rc
total 36
drwxr-xr-x   2 root root  4096 Aug 23 07:28 .
drwxr-xr-x 283 root root 12288 Sep  5 18:11 ..
-rw-r--r--   1 root root  7110 Jul 12 03:05 nvidia-application-profiles-470.57.02-key-documentation
-rw-r--r--   1 root root  9082 Jul 12 03:05 nvidia-application-profiles-470.57.02-rc
  1. Take note of the registry key file that got installed. It will have some numbers in the name ending with"key-documentation"
  2. Create a symlink to the actual key. ln -s <actual> <expected> For example on my system
$ sudo ln -s nvidia-application-profiles-470.57.02-key-documentation nvidia-application-profiles-key-documentation
1 Like