Check manual page of nvidia.temp

nvidia.temp

Author: Mathias Kettner

License: GPL

Distribution: official part of Check_MK

Supported Agents: Linux

This checks monitors the temperatures of the GPU Core and ambient temperatur of an NVIDIA graphics card using the command line tool nvidia-settings. The check only works if that tool is installed and a X server is running on display :0.

Item

The name of the sensor as reported by nvidia-settings -q all

Check parameters

Parameter	Type	Description
warning	int	ambient temperature at which a warning level is triggered
critical	int	critical level for ambient temperature
warning	int	GPU core temperature at which a warning level is triggered
critical	int	critical level for GPU core temperature

Performance data

One variable: the current temperature in degrees celsius.

Inventory

All available temperature sensors are automatically inventorized.

Configuration variables

Variable	Type	Description
nvidia_temp_default_levels	int, int, int, int	Default levels for warning and critical. This is preset to (60, 65, 90, 95).

Examples

# set default levels for ambient and GPU sensors
nvidia_temp_default_levels = (50, 60, 70, 80)