CPU and HDD temperature logging system

Use case

You might run into a situation where your old laptop or your new work machine starts running hotter even when you are running no task which is either CPU or I/O intensive in particular. Maybe it’s time to do a clean install of your favourite linux distro (the lazy way) or deep dive and investigate what might be causing the overheating (the time-consuming way).

Either way you want to see and be able to analyze how hot your CPU cores and your hard disk is running at. A record of these temperatures persisting over usage sessions might help narrow down the cause of the heating.

My old laptop which was acting as a media server started overheating. Thus I decided to write this script.

Problem definition

Write a script that logs CPU core temperature and Hard Disk Drive (HDD) temperatures every 30 seconds in a file. The script must start automatically on booting.

Solution

For demonstration purposes I am going to use Ubuntu throughout this guide.

First let us install some dependencies.

sudo apt-get install lm-sensors
sudo apt-get install hddtemp
sudo chmod u+s /usr/sbin/hddtemp

The third command lets us run hddtemp without using sudo.

Now let us try running these commands on the terminal.

abhishek@ThinkPad-P50:~$ sensors
acpitz-virtual-0
Adapter: Virtual device
temp1:        +43.0°C  (crit = +128.0°C)

thinkpad-isa-0000
Adapter: ISA adapter
fan1:           0 RPM

coretemp-isa-0000
Adapter: ISA adapter
Physical id 0:  +43.0°C  (high = +100.0°C, crit = +100.0°C)
Core 0:         +41.0°C  (high = +100.0°C, crit = +100.0°C)
Core 1:         +41.0°C  (high = +100.0°C, crit = +100.0°C)
Core 2:         +39.0°C  (high = +100.0°C, crit = +100.0°C)
Core 3:         +40.0°C  (high = +100.0°C, crit = +100.0°C)

abhishek@ThinkPad-P50:~$ hddtemp /dev/sda1
/dev/sda1: ST500LM021-1KJ152: 43°C

Now we will extract the CPU core temperatures from the ouput of sensors command using pipe

sensors | awk  '/Core/{print $3}' ORS=' '

Output

abhishek@ThinkPad-P50:~$ sensors | awk  '/Core/{print $3}' ORS=' '
+40.0°C +41.0°C +39.0°C +40.0°C abhishek@ThinkPad-P50:~$ 

” | “ is known as unnamed pipe and is used to pass output of command on the left as input to command on the right.

” awk “” is pattern scanning and processing language. Check here for understanding what each option does and here for a complete tuorial.

Similarly we can extract HDD tempretures using

hddtemp /dev/sda | awk  '//{print $3}

Output

abhishek@ThinkPad-P50:~$ hddtemp /dev/sda | awk  '//{print $3}'
41°C

You can find temperatures for other hard disks by replacing /dev/sda in the command with other appropriate names. You can find list of devices using

lsblk

Now that we have extracted the values that we need, we now

echo -e "$(date) \n " >> ~/log
   while :
   do
		core_temp=`sensors | awk  '/Core/{print $3}' ORS=' '`
        hdd_temp=`hddtemp /dev/sda | awk  '//{print $4}'`
        echo $core_temp $hdd_temp >> ~/log
		echo  -e "\n"
        sleep 30
done

We save the date and time each time the script is started. After that we go in an infinite loop which gathers required data and appends to “~/log” ie. in a file named “log” in your home directory every 30 seconds.

The file should look like this

Fri Jul 21 15:23:23 IST 2017 
 
+40.0°C +41.0°C +38.0°C +40.0°C 42°C
+45.0°C +42.0°C +48.0°C +42.0°C 42°C
+42.0°C +43.0°C +40.0°C +41.0°C 42°C

Now we want this script to automatically start and stop as the systems starts and stops. So we add the header as specified in the Linux Standard Base(LSB)

update-rc.d

#!/bin/bash
### BEGIN INIT INFO
# Provides:          temperature_recording_daemon
# Required-Start:
# Required-Stop:
# Default-Start:     2 3 4 5
# Default-Stop:      0 1 6
# Short-Description: Record CPU and HDD temperatures
# Description:       This script stores CPU and HDD temperatures every 30 seconds in the format
#                    CORE1_temp CORE2_temp HDD_temp
#
### END INIT INFO

case "$1" in
 start)
   echo -e "$(date) \n " >> /home/log
   while :
   do
	core_temp=`sensors | awk  '/Core/{print $3}' ORS=' '`
        hdd_temp=`hddtemp /dev/sda | awk  '//{print $3}'`
        echo $core_temp $hdd_temp >> /home/log
	echo  -e "\n"
        sleep 30
   done
   ;;
 stop)

   ;;
 restart)

   ;;
 *)
   ;;
esac

This page helps understand the header. We then move the script to /etc/init.d/, make it executable by adding the +x flag.

sudo cp <name> /etc/int.d/<name>
sudo chmod 755 /etc/init.d/<name>
sudo chown root:root /etc/init.d/<name>
sudo update-rc.d <name> defaults
sudo update-rc.d <name> enable

And voila! Your script is ready and should start automatically when you boot into the system.