lm sensors

From ArchWiki
Revision as of 23:17, 11 September 2012 by MrBougo (Talk | contribs) (fix automatic deployment commands)

Jump to: navigation, search

zh-CN:Lm sensors lm_sensors (Linux monitoring sensors) is a free and open-source application that provides tools and drivers for monitoring temperatures, voltage, and fans. This document explains how to install, configure, and use lm_sensors so that you can monitor CPU temperatures, motherboard temperatures, and fan speeds.



Install the lm_sensors package from the official repositories.

Setting up lm_sensors

Use sensors-detect to detect and generate a list of kernel modules:

# sensors-detect

This will create the /etc/conf.d/lm_sensors configuration file which is used by the sensors daemon to automatically load kernel modules on boot. You will be asked if you want to probe for various hardware. The "safe" answers are the defaults, so just hitting Template:Keypress to all the questions will generally not cause any problems.

When the detection is finished, you will be presented with a summary of the probes. Here is an example summary from my system:

# sensors-detect
Now follows a summary of the probes I have just done.
Just press ENTER to continue:
Driver `it87':
  * ISA bus, address 0x290
     Chip `ITE IT8718F Super IO Sensors' (confidence: 9)
Driver `coretemp':
  * Chip `Intel Core family thermal sensor' (confidence: 9)

If you plan on using the daemon, be sure to answer YES when asked if you want to to generate /etc/conf.d/lm_sensors.

To automatically load the kernel modules at boot time, add sensors to the DAEMONS array in /etc/rc.conf

DAEMONS=(syslog-ng crond ... sensors ...)

On a systemd system you just need:

systemctl enable lm_sensors.service

Alternatively, instead of using the daemon, you can add the modules to the MODULES array in /etc//etc/modules-load.d/lm_sensors.conf:


Automatic lm_sensors deployment

If you wish to deploy lm-sensors on multiple diferent linux machines issue is that sensors-detect ask you quite a few questions. There are few tricks that you can use to automate replies.

First one is if you wish to accept defaults which sensors-detect suggest you need just to press [ENTER] all the time. To automate this use this one liner:

# yes "" | sensors-detect

If you wish to override defaults and answer YES to all questions then use this oneliner:

# yes | sensors-detect

Testing your lm_sensors

To test your setup, load the kernel modules manually or by using the /etc/rc.d/sensors init script. You do NOT have to do both. Example manually adding them

# modprobe it87
# modprobe coretemp

Example using the script

# rc.d start sensors

You should see something like this when you run sensors

$ sensors
Adapter: ISA adapter
Core 0:      +30.0°C  (high = +76.0°C, crit = +100.0°C)  

Adapter: ISA adapter
Core 1:      +30.0°C  (high = +76.0°C, crit = +100.0°C)  

Adapter: ISA adapter
Core 2:      +32.0°C  (high = +76.0°C, crit = +100.0°C)  

Adapter: ISA adapter
Core 3:      +30.0°C  (high = +76.0°C, crit = +100.0°C)  

Adapter: ISA adapter
in0:         +1.17 V  (min =  +0.00 V, max =  +4.08 V)   
in1:         +1.31 V  (min =  +1.28 V, max =  +1.68 V)   
in2:         +3.28 V  (min =  +2.78 V, max =  +3.78 V)   
in3:         +2.88 V  (min =  +2.67 V, max =  +3.26 V)   
in4:         +2.98 V  (min =  +2.50 V, max =  +3.49 V)   
in5:         +1.34 V  (min =  +0.58 V, max =  +1.34 V)   ALARM
in6:         +2.02 V  (min =  +1.04 V, max =  +1.36 V)   ALARM
in7:         +2.83 V  (min =  +2.67 V, max =  +3.26 V)   
Vbat:        +3.28 V
fan1:       1500 RPM  (min = 3245 RPM)  ALARM
fan2:          0 RPM  (min = 3245 RPM)  ALARM
fan3:          0 RPM  (min = 3245 RPM)  ALARM
temp1:       +18.0°C  (low  = +127.0°C, high = +64.0°C)  sensor = thermal diode
temp2:       +32.0°C  (low  = +127.0°C, high = +64.0°C)  sensor = thermistor
temp3:       +38.0°C  (low  = +127.0°C, high = +64.0°C)  sensor = thermistor
cpu0_vid:   +2.050 V

Adapter: Virtual device
temp1:       +18.0°C  (crit = +64.0°C)

Reading SPD values from memory modules (optional)

To read the SPD timing values from your memory modules, install i2c-tools from the official repositories. Once you have i2c-tools installed, you will need to load the eeprom kernel module.

# modprobe eeprom

Finally, you can view your memory information with decode-dimms.

Here is partial output from one machine:

$ decode-dimms
# decode-dimms version 5733 (2009-06-09 13:13:41 +0200)

Memory Serial Presence Detect Decoder
By Philip Edelbrock, Christian Zuckschwerdt, Burkart Lingner,
Jean Delvare, Trent Piepho and others

Decoding EEPROM: /sys/bus/i2c/drivers/eeprom/0-0050
Guessing DIMM is in                             bank 1

---=== SPD EEPROM Information ===---
EEPROM CRC of bytes 0-116                       OK (0x583F)
# of bytes written to SDRAM EEPROM              176
Total number of bytes in EEPROM                 512
Fundamental Memory type                         DDR3 SDRAM
Module Type                                     UDIMM

---=== Memory Characteristics ===---
Fine time base                                  2.500 ps
Medium time base                                0.125 ns
Maximum module speed                            1066MHz (PC3-8533)
Size                                            2048 MB
Banks x Rows x Columns x Bits                   8 x 14 x 10 x 64
Ranks                                           2
SDRAM Device Width                              8 bits
tCL-tRCD-tRP-tRAS                               7-7-7-33
Supported CAS Latencies (tCL)                   8T, 7T, 6T, 5T

---=== Timing Parameters ===---
Minimum Write Recovery time (tWR)               15.000 ns
Minimum Row Active to Row Active Delay (tRRD)   7.500 ns
Minimum Active to Auto-Refresh Delay (tRC)      49.500 ns
Minimum Recovery Delay (tRFC)                   110.000 ns
Minimum Write to Read CMD Delay (tWTR)          7.500 ns
Minimum Read to Pre-charge CMD Delay (tRTP)     7.500 ns
Minimum Four Activate Window Delay (tFAW)       30.000 ns

---=== Optional Features ===---
Operable voltages                               1.5V
RZQ/6 supported?                                Yes
RZQ/7 supported?                                Yes
DLL-Off Mode supported?                         No
Operating temperature range                     0-85C
Refresh Rate in extended temp range             1X
Auto Self-Refresh?                              Yes
On-Die Thermal Sensor readout?                  No
Partial Array Self-Refresh?                     No
Thermal Sensor Accuracy                         Not implemented
SDRAM Device Type                               Standard Monolithic

---=== Physical Characteristics ===---
Module Height (mm)                              15
Module Thickness (mm)                           1 front, 1 back
Module Width (mm)                               133.5
Module Reference Card                           B

---=== Manufacturer Data ===---
Module Manufacturer                             Invalid
Manufacturing Location Code                     0x02
Part Number                                     OCZ3G1600LV2G     


Using sensor data

Graphical Front-ends

There are a variety of front-ends for sensors data.

  • xsensors - X11 interface to lm_sensors
  • xfce4-sensors-plugin - A lm_sensors plugin for the Xfce panel
  • conky - Conky is an advanced, highly configurable system monitor for X based on torsmo
  • kdeutils-superkaramba - Superkaramba is a tool which gives posibility to create different widgets for KDE desktop. Check the karamba section on kde-look.org for examples of making karamba front-ends for sensors data.
  • sensors-applet - applet for the GNOME Panel to display readings from hardware sensors, including CPU temperature, fan speeds and voltage readings.


There is an optional daemon called sensord (included with the lm_sensors package) which can log your data to a round robin database (rrd) and later visualize graphically. See the sensord man page for details.


Renumbering Cores for Multi-CPU Systems

In rare cases, the actual numbering of physical cores on multi-processor motherboards can be incorrect. Consider the following HP Z600 workstation with dual Xeons:

$ sensors
Adapter: ISA adapter
Core 0:       +65.0°C  (high = +85.0°C, crit = +95.0°C)
Core 1:       +65.0°C  (high = +85.0°C, crit = +95.0°C)
Core 9:       +66.0°C  (high = +85.0°C, crit = +95.0°C)
Core 10:      +66.0°C  (high = +85.0°C, crit = +95.0°C)

Adapter: ISA adapter
Core 0:       +54.0°C  (high = +85.0°C, crit = +95.0°C)
Core 1:       +56.0°C  (high = +85.0°C, crit = +95.0°C)
Core 9:       +60.0°C  (high = +85.0°C, crit = +95.0°C)
Core 10:      +61.0°C  (high = +85.0°C, crit = +95.0°C)

Adapter: ISA adapter
fan1:        1730 RPM
fan2:        1746 RPM
fan3:        1224 RPM
fan4:        2825 RPM
temp1:        +46.0°C
temp2:        +37.0°C
temp3:        +23.0°C
temp4:       -128.0°C

Note the cores are numbered 0, 1, 9, 10 which is repeated into the second CPU. Most users want the core temperatures to report out in sequential order, i.e. 0,1,2,3,4,5,6,7. Fixing the order is accomplished in two steps.

Step 1. ID what each chip is reporting

Run sensors with the -u switch to see what options are available for each physical chip:

$ sensors -u coretemp-isa-0000
Adapter: ISA adapter
Core 0:
  temp2_input: 61.000
  temp2_max: 85.000
  temp2_crit: 95.000
  temp2_crit_alarm: 0.000
Core 1:
  temp3_input: 61.000
  temp3_max: 85.000
  temp3_crit: 95.000
  temp3_crit_alarm: 0.000
Core 9:
  temp11_input: 62.000
  temp11_max: 85.000
  temp11_crit: 95.000
Core 10:
  temp12_input: 63.000
  temp12_max: 85.000
  temp12_crit: 95.000
$ sensors -u coretemp-isa-0004
Adapter: ISA adapter
Core 0:
  temp2_input: 53.000
  temp2_max: 85.000
  temp2_crit: 95.000
  temp2_crit_alarm: 0.000
Core 1:
  temp3_input: 54.000
  temp3_max: 85.000
  temp3_crit: 95.000
  temp3_crit_alarm: 0.000
Core 9:
  temp11_input: 59.000
  temp11_max: 85.000
  temp11_crit: 95.000
Core 10:
  temp12_input: 59.000
  temp12_max: 85.000
  temp12_crit: 95.000

Step 2. Redefine the cores

Create /etc/sensors.d/cores.conf wherein the new definitions are defined based on the output of step 1:

chip "coretemp-isa-0000"

    label temp2 "Core 0"
    label temp3 "Core 1"
    label temp11 "Core 2"
    label temp12 "Core 3"

chip "coretemp-isa-0004"

    label temp2 "Core 4"
    label temp3 "Core 5"
    label temp11 "Core 6"
    label temp12 "Core 7"

Problem solved. Output after completing these steps:

$ sensors
Adapter: ISA adapter
Core0:        +64.0°C  (high = +85.0°C, crit = +95.0°C)
Core1:        +63.0°C  (high = +85.0°C, crit = +95.0°C)
Core2:        +65.0°C  (high = +85.0°C, crit = +95.0°C)
Core3:        +66.0°C  (high = +85.0°C, crit = +95.0°C)

Adapter: ISA adapter
Core4:        +53.0°C  (high = +85.0°C, crit = +95.0°C)
Core5:        +54.0°C  (high = +85.0°C, crit = +95.0°C)
Core6:        +59.0°C  (high = +85.0°C, crit = +95.0°C)
Core7:        +60.0°C  (high = +85.0°C, crit = +95.0°C)

Adapter: ISA adapter
fan1:        1734 RPM
fan2:        1726 RPM
fan3:        1222 RPM
fan4:        2827 RPM
temp1:        +45.0°C  
temp2:        +37.0°C  
temp3:        +23.0°C  
temp4:       -128.0°C  

Sensors not working since Linux 2.6.31

A change in version 2.6.31 has made some sensors stop working. See this FAQ entry for a detailed explanation and for some example errors. To fix sensors, add the following to your kernel boot line (e.g. in your GRUB/GRUB2 configuration file) and reboot your machine:

Warning: In some situations, this may be dangerous. Consult the FAQ for details.

Note that in most cases the information is still accessible via other modules (e.g. via ACPI modules) for the hardware in question. Many utilities and monitors (e.g. /usr/bin/sensors) can gather information from either source. Where possible, this is the preferred solution.

K10Temp Module

Some K10 processors have issues with their temperature sensor. From the kernel documentation (linux-<version>/Documentation/hwmon/k10temp):

All these processors have a sensor, but on those for Socket F or AM2+, the sensor may return inconsistent values (erratum 319). The driver will refuse to load on these revisions unless you specify the force=1 module parameter.
Due to technical reasons, the driver can detect only the mainboard's socket type, not the processor's actual capabilities. Therefore, if you are using an AM3 processor on an AM2+ mainboard, you can safely use the force=1 parameter.

On affected machines the module will report "unreliable CPU thermal sensor; monitoring disabled". If you still want to use the module you can:

# rmmod k10temp
# modprobe k10temp force=1

Confirm with Lm_sensors#Testing your lm_sensors that the sensor is in fact valid and reliable. If it is, you can edit /etc/modprobe.d/k10temp.conf and add:

options k10temp force=1

This will allow the module to load at boot.

See also

  • hddtemp - Software to read temperatures of hard drives.
  • monitorix - Monitorix is a free, open source, lightweight system monitoring tool designed to monitor as many services and system resources as possible.