Here is a custom script that reports the nvidia-smi temperature of the Nvidia cards in your system using NRPE. Conversely,  NRPE performs much better than NCPA and my protocol of choice.

*Note: Install NRPE to your Linux box first, and create this file (gpu_temp.sh) in your /usr/lib/nagios/plugins directory for Nagios Core 4+

Nagios NRPE Query Command

check_nrpe!gpu_temp.sh

This will only query the first GPU that is found via nvidia-smi CLI command. It will do warnings and alerts at the given thresholds.

nvidia smi nagios

The gpu_temp.sh Script

Use this script to query nvidia-smi, thus reporting the temp of the card.

############################################
## Sample nvidia commands to query gpu
##
##nvidia-smi -q -d TEMPERATURE | grep "GPU Current Temp" | awk '{print $5}'
##nvidia-smi --query-gpu=temperature.gpu --format=csv,noheader
##nvidia-smi | grep '[0-9][0-9]C' | awk '{print $3}' | sed 's/C//'
############################################`

Here is the script code:

#!/bin/bash
gpu_temp=`nvidia-smi | grep '[0-9][0-9]C' | awk '{print $3}'`
case $gpu_temp in
[1-70]*)
echo "OK - $gpu_temp in normal operating range."
exit 0
;;
[71-80]*)
echo "WARNING - $gpu_temp in high operating range."
exit 1
;;
[81-100]*)
echo "CRITICAL - $gpu_temp in extreme operating range. Shutdown ASAP!"
exit 2
;;
*)
echo "UNKNOWN - $gpu_temp is current temperature."
exit 3
;;
esac`

You will need to modify this as the nvidia-smi tool will be updated in the future and the grep command will have to be reconfigured if due to any output changes.