jackbenny / check_gputemp

A Nagios plugin to monitor the GPU temperature of ATI based graphics card

• Lifelong learner 🏫 📔
• Linux consultant 🐧 🐳
• Author 📚 📙
• Publisher 📰 📗

Books I've written

• Grunderna i programmering (Swedish)
  • All code examples (3rd edition)
  • All code examples (2nd edition)
• C-programmering i Linux, macOS, BSD och Solaris (Swedish)
  • All code examples
• Linux systemadministration från grunden (Swedish)
  • All code examples
• Linux System Programming Techniques (English)
  • All code examples
• Docker från grunden (Swedish)
  • All code files
• Ansible från grunden (Swedish)
  • All code files
• Demonerna på internet (Swedish)
  • All code files

Courses I've made

• Linux Command Line: From novice to wizard (English)
• Hands-On Linux System Administration (English)

check_gputemp's Issues

running as standard user, root and nagios user?

Hi,

Thank you for this plugin. It works, but I have the following issue under Ubuntu 14. I'm using the system as "myuser". When I'm logged in as "myuser" I can run aticonfig and it returns proper value.

If I run aticonfig as root user, I get:

No protocol specified
ERROR - X needs to be running to perform AMD Overdrive(TM) commands

Therefore I use:

sudo -u myuser aticonfig --odgt

And I get OK result:

Default Adapter - AMD Radeon HD 7900 Series
Sensor 0: Temperature - 62.00 C

And the check is OK when running as root user:

GPU 0 OK - Temperature is 62 | Sensor 0: Temperature - 62.00 C

So I modified the script and I added "sudo -u myuser" when aticonfig command is called. But when I run the check from remote Nagios server via NRPE I now get this result:

"GPU 0 OK - Temperature is | "

So it seems that aticonfig command is not properly executed when called from remote Nagios server via NRPE. But I see this in my local machine (where GPUs are):

nrpe[8841]: Running command: /usr/lib/nagios/plugins/check_amdgputemp.sh --adapter 0 -w 68 -c 75
Command completed with return code 0 and output: GPU 0 OK - Temperature is |
Return Code: 0, Output: GPU 0 OK - Temperature is |

What is the proper way to configure this so that Nagios from remote server would execute amdgup checks properly?

Thanks!