Hardware monitoring
Author | Damien Martins |
---|---|
Compatibility | Xymon 4.2.2/4.3.12 |
Requirements | sh (or bash), hddtemp, smartmontools |
Download | Part of https://github.com/doktoil-makresh/xymon-plugins.git |
Last Update | 2022-07-13 |
Description
A shell script to monitor hardware sensors (hdd and motherboard actually).
Installation
Client side
Untar this package, put hobbit-hardware.sh in $XYMONCLIENTHOME/ext directory Put xymon-hardware.cfg in $XYMONCLIENTHOME/etc directory Modify variables in both files to fit your needs/system User 'xymon' should be allowed to use sudo on some commands (check variables including 'sudo' in xymon-hardware.sh)
Server side
Add hardware to you $XYMONHOME/server/hosts line for the host running this script
Source
hobbit-hardware.sh
Show Code ⇲
Hide Code ⇱
#!/bin/bash # ALL THIS SCRIPT IS UNDER GPL LICENSE # Version 0.6 # Title: xymon-hardware # Author: Damien Martins ( doctor |at| makelofine |dot| org) # Date: 2022-07-13 # Purpose: Check Uni* hardware sensors # Platforms: Uni* having lm-sensor and hddtemp utilities # Tested: Xymon 4.3.4 / hddtemp version 0.3-beta15 (Debian Lenny and Etch packages) / sensors version 3.0.2 with libsensors version 3.0.2 (Debian Lenny package) / sensors version 3.0.1 with libsensors version 3.0.1 (Debian Etch package) ################################################################################# # YOU MUST CONFIGURE LM-SENSORS IN ORDER TO GET VALUES BEFORE USING THIS SCRIPT # ################################################################################# #This script should be stored in ext directory, located in Xymon/Xymon client home (typically ~xymon/client/ext or ~xymon/client/ext). #You must configure the xymon-hardware.cfg file (or whatever name defined in CONFIG_FILE) #Debug if [ "$1" == "debug" ] ; then echo "Debug ON" BB=echo XYMONCLIENTHOME="/usr/local/Xymon/client/" XYMONTMP="$PWD" BBDISP=your_xymon_server MACHINE=$(hostname) CAT="/bin/cat" AWK="/usr/bin/nawk" GREP="/bin/grep" RM="/bin/rm" CUT="/usr/bin/cut" DATE="/bin/date" SED="/bin/sed" CONFIG_FILE="xymon-hardware.cfg" TMP_FILE="xymon-hardware.tmp" MSG_FILE="xymon-hardware.msg" fi #Change to fit your system/wills : TEST="hardware" MSG_FILE="${XYMONTMP}/xymon-hardware.msg" CONFIG_FILE="${XYMONCLIENTHOME}/etc/xymon-hardware.cfg" TMP_FILE="${XYMONTMP}/xymon-hardware.tmp" CMD_HDDTEMP="sudo /usr/sbin/hddtemp" SENSORS="/usr/bin/sensors" BC="/usr/bin/bc" SMARTCTL="sudo /usr/sbin/smartctl" OMREPORT="/opt/dell/srvadmin/sbin/omreport" HPACUCLI="sudo /usr/sbin/hpacucli" #Don't change anything from here (or assume all responsibility) unset YELLOW unset RED #Basic tests : if [ -z "$XYMONCLIENTHOME" ] ; then echo "XYMONCLIENTHOME not defined !" exit 1 fi if [ -z "$XYMONTMP" ] ; then echo "XYMONTMP not defined !" exit 1 fi if [ -z "$BB" ] ; then echo "BB not defined !" exit 1 fi if [ -z "$BBDISP" ] ; then echo "BBDISP not defined !" exit 1 fi if [ -z "$MACHINE" ] ; then echo "MACHINE not defined !" exit 1 fi #Let's start baby !!! # #Hard disk temperature monitoring if [ -f "$MSG_FILE" ] ; then "$RM" "$MSG_FILE" fi DISK_WARNING_TEMP=$($GREP ^DISK_WARNING_TEMP= $CONFIG_FILE | $SED s/^DISK_WARNING_TEMP=//) DISK_PANIC_TEMP=$($GREP ^DISK_PANIC_TEMP= $CONFIG_FILE | $SED s/^DISK_PANIC_TEMP=//) function use_hddtemp () { for DISK in $("$GREP" "^DISK=" "$CONFIG_FILE" | "$SED" s/^DISK=//) ; do HDD_TEMP="$($CMD_HDDTEMP $DISK | $SED s/..$// | $AWK '{print $NF}')" if [ ! "$(echo $HDD_TEMP | grep "^[ [:digit:] ]*$")" ] ; then RED=1 LINE="&red Disk $DISK temperature is UNKNOWN (HDD_TEMP VALUE IS : $HDD_TEMP). It seems S.M.A.R.T. is no more responding !!!" echo "La temp�rature de $DISK n'est pas un nombre :/ HDD_TEMP : $HDD_TEMP" elif [ "$HDD_TEMP" -ge "$DISK_PANIC_TEMP" ] ; then RED=1 LINE="&red Disk temperature is CRITICAL (Panic is $DISK_PANIC_TEMP) : "$DISK"_temperature: ${HDD_TEMP}" elif [ "$HDD_TEMP" -ge "$DISK_WARNING_TEMP" ] ; then YELLOW="1" LINE="&yellow Disk temperature is HIGH (Warning is $DISK_WARNING_TEMP) : "$DISK"_temperature: ${HDD_TEMP}" elif [ "$HDD_TEMP" -lt "$DISK_WARNING_TEMP" ] ; then LINE="&green Disk temperature is OK (Warning is $DISK_WARNING_TEMP) : "$DISK"_temperature: ${HDD_TEMP}" fi echo "$LINE" >> "$MSG_FILE" done } function use_smartctl () { SMARTCTL_CHIPSET="$($GREP ^SMARTCTL_CHIPSET= $CONFIG_FILE | $SED s/^SMARTCTL_CHIPSET=//)" if [ $SMARTCTL_CHIPSET ] ; then SMARTCTL_ARGS="-A -d $SMARTCTL_CHIPSET" else SMARTCTL_ARGS="-A" fi for DISK in $("$GREP" "^DISK=" "$CONFIG_FILE" | "$SED" s/^DISK=//) ; do HDD_TEMP="$($SMARTCTL $SMARTCTL_ARGS $DISK | $GREP "^194" | $AWK '{print $10}')" if [ ! "$(echo $HDD_TEMP | grep "^[ [:digit:] ]*$")" ] ; then RED=1 LINE="&red Disk $DISK temperature is UNKNOWN (HDD_TEMP VALUE IS : $HDD_TEMP). It seems S.M.A.R.T. is no more responding !!!" echo "La temp�rature de $DISK n'est pas un nombre :/ HDD_TEMP : $HDD_TEMP" elif [ "$HDD_TEMP" -ge "$DISK_PANIC_TEMP" ] ; then RED=1 LINE="&red Disk temperature is CRITICAL (Panic is $DISK_PANIC_TEMP) : "$DISK"_temperature: ${HDD_TEMP}" elif [ "$HDD_TEMP" -ge "$DISK_WARNING_TEMP" ] ; then YELLOW="1" LINE="&yellow Disk temperature is HIGH (Warning is $DISK_WARNING_TEMP) : "$DISK"_temperature: ${HDD_TEMP}" elif [ "$HDD_TEMP" -lt "$DISK_WARNING_TEMP" ] ; then LINE="&green Disk temperature is OK (Warning is $DISK_WARNING_TEMP) : "$DISK"_temperature: ${HDD_TEMP}" fi echo "$LINE" >> "$MSG_FILE" done } #Motherboard sensors monitoring (CPU, Mobo, Fans...) function test_temperature () { SOURCE=$1 TEMPERATURE=$2 WARNING=$3 PANIC=$4 #echo "Source : $SOURCE #Temperature : $TEMPERATURE #Warning : $WARNING #Panic : $PANIC" if [ $(echo "$TEMPERATURE >= $PANIC" | "$BC") -eq 1 ] ; then RED=1 LINE="&red $SOURCE temperature is CRITICAL !!! (Panic is $PANIC) : "$SOURCE"_temperature: $TEMPERATURE" elif [ $(echo "$TEMPERATURE >= $WARNING" | "$BC") -eq 1 ] ; then YELLOW=1 LINE="&yellow $SOURCE temperature is HIGH ! (Warning is $WARNING) : "$SOURCE"_temperature: $TEMPERATURE" elif [ $(echo "$TEMPERATURE < $WARNING" | "$BC") -eq 1 ] ; then LINE="&green $SOURCE temperature is OK (Warning is $WARNING) : "$SOURCE"_temperature: $TEMPERATURE" fi echo "$LINE" >> "$MSG_FILE" unset MIN MAX PANIC VALUE WARNING } function test_fan () { SOURCE=$1 RPM=$2 MIN=$3 #echo "Source : $SOURCE #RPM : $RPM #MIN : $MIN" if [ $(echo "$RPM <= $MIN" |"$BC") -eq 1 ] ; then RED=1 LINE="&red $SOURCE RPM speed is critical !!! (Lower or equal to $MIN) : "$SOURCE"_rpm: $RPM" elif [ $(echo "$RPM > $MIN" |"$BC") -eq 1 ] ; then LINE="&green $SOURCE RPM is OK (Higher than $MIN) : "$SOURCE"_rpm: $RPM" fi echo "$LINE" >> "$MSG_FILE" unset MIN MAX PANIC VALUE WARNING } function test_volt () { SOURCE=$1 VOLT=$2 MIN=$3 MAX=$4 #echo "Source : $SOURCE #Volt : $VOLT #Min : $MIN #Max : $MAX" if [ $(echo "$VOLT < $MIN" | "$BC") -eq 1 ] || [ $(echo "$VOLT > $MAX" | "$BC") -eq 1 ] ; then RED=1 LINE="&red $SOURCE voltage is OUT OF RANGE !!! (between $MIN and $MAX) : "$SOURCE"_volt: $VOLT" elif [ $(echo "$VOLT == $MIN" |"$BC") -eq 1 ] || [ $(echo "$VOLT == $MAX" |"$BC") -eq 1 ] ; then YELLOW="1" LINE="&yellow $SOURCE voltage is very NEAR OF LIMITS ! (between $MIN and $MAX) : "$SOURCE"_volt: $VOLT" elif [ $(echo "$VOLT > $MIN" |"$BC") -eq 1 ] && [ $(echo "$VOLT < $MAX" |"$BC") -eq 1 ] ; then LINE="&green $SOURCE voltage is OK (between $MIN and $MAX) : "$SOURCE"_volt: $VOLT" fi echo "$LINE" >> "$MSG_FILE" unset MIN MAX PANIC VALUE WARNING } function find_type () { LINE=$1 echo "$LINE" | "$GREP" -q "in[0-9]" if [ $? -eq 0 ] ; then TYPE=volt else echo "$LINE" | "$GREP" -q "fan[0-9]" if [ $? -eq 0 ] ; then TYPE=fan else echo "$LINE" |"$GREP" -q "temp[0-9]" if [ $? -eq 0 ] ; then TYPE=temp fi fi fi # echo "Type : $TYPE" } function use_lmsensors () { SENSOR_PROBE="$($GREP ^SENSOR_PROBE= $CONFIG_FILE | $SED s/^SENSOR_PROBE=//)" if [ -z $SENSOR_PROBE ] ; then echo "No sensor probe configured" break fi "$SENSORS" -uA "$SENSOR_PROBE" | "$GREP" : | "$GREP" -v beep_enable | $GREP -v "alarm" | $GREP -v "type" > "$TMP_FILE" while read SENSORS_LINE ; do #echo "Ligne : $SENSORS_LINE" echo $SENSORS_LINE | "$AWK" -F: '{print $2}' | "$GREP" -q "[0-9]" if [ $? -ne 0 ] ; then TITLE=$(echo $SENSORS_LINE | "$AWK" -F: '{print $1}' | $SED 's/\ /_/g' |$SED 's/^-/Negative_/' |$SED 's/^+/Positive_/') # echo "Title : $TITLE" else find_type "$SENSORS_LINE" echo $SENSORS_LINE | "$GREP" -q "input:" if [ $? -eq 0 ] ; then VALUE=$(echo $SENSORS_LINE | "$AWK" '{print $2}') # echo "Value : $VALUE" fi echo $SENSORS_LINE |"$GREP" -q "_max:" if [ $? -eq 0 ] ; then PANIC=$(echo $SENSORS_LINE | "$AWK" '{print $2}') MAX=$PANIC # echo "Panic : $PANIC" fi echo $SENSORS_LINE |"$GREP" -q "_max_hyst:" if [ $? -eq 0 ] ; then WARNING=$(echo $SENSORS_LINE | "$AWK" '{print $2}') # echo "Warning : $WARNING" fi echo $SENSORS_LINE |"$GREP" -q "_min:" if [ $? -eq 0 ] ; then MIN=$(echo $SENSORS_LINE | "$AWK" '{print $2}') # echo "Min : $MIN" fi if [ "$TYPE" == "volt" ] && [ "$MIN" ] && [ $VALUE ] && [ $MAX ] ; then test_volt $TITLE $VALUE $MIN $MAX elif [ "$TYPE" == "fan" ] && [ $TITLE ] && [ $MIN ] && [ $VALUE ] ; then test_fan $TITLE $VALUE $MIN elif [ "$TYPE" == "temp" ] && [ $TITLE ] && [ $VALUE ] && [ $WARNING ] && [ $PANIC ] ; then test_temperature $TITLE $VALUE $WARNING $PANIC fi fi done < "$TMP_FILE" } function use_openmanage () { rm -f ${XYMONTMP}/xymon-hardware_volts.tmp ${XYMONTMP}/xymon-hardware_fans.tmp ${XYMONTMP}/xymon-hardware_disks.tmp #Tests temperatures : CHASSIS_TEMP=$($OMREPORT chassis temps | grep Reading |awk '{print $3}' | $AWK -F\. '{print $1}') CHASSIS_TEMP_WARNING=$($OMREPORT chassis temps | grep "Maximum Warning Threshold" |awk '{print $5}' | $AWK -F\. '{print $1}' ) CHASSIS_TEMP_ALERT=$($OMREPORT chassis temps | grep "Maximum Failure Threshold" |awk '{print $5}' | $AWK -F\. '{print $1}') if [ $CHASSIS_TEMP_WARNING -ge $CHASSIS_TEMP_ALERT ] ; then echo "Erreur, la valeur CHASSIS_TEMP_WARNING est superieure ou egale a CHASSIS_TEMP_ALERT !!!" exit 2 fi if [ $CHASSIS_TEMP -ge $CHASSIS_TEMP_ALERT ] ; then CHASSIS_TEMP_STATUS=red echo "&red La temperature du chassis est en ALERTE !!! : temperature_chassis: $CHASSIS_TEMP" >> $MSG_FILE RED=1 elif [ $CHASSIS_TEMP -ge $CHASSIS_TEMP_WARNING ] ; then CHASSIS_TEMP_STATUS=yellow YELLOW=1 echo "&yellow La temperature du chassis est en LIMITE-LIMITE !!! : temperature_chassis: $CHASSIS_TEMP" >> $MSG_FILE elif [ $CHASSIS_TEMP -lt $CHASSIS_TEMP_WARNING ] ; then CHASSIS_TEMP_STATUS=green echo "&green Les voltages sont Ok !" >> $MSG_FILE else echo "Erreur dans les valeurs de temperatures : CHASSIS_TEMP : $CHASSIS_TEMP CHASSIS_TEMP_WARNING : $CHASSIS_TEMP_WARNING CHASSIS_TEMP_ALERT : $CHASSIS_TEMP_ALERT" exit 2 fi #Tests voltages : $OMREPORT chassis volts |grep Health | awk ' {print $3}' | grep -q Ok if [ $? -eq 0 ] ; then VOLT_GLOBAL_STATUS=green else $OMREPORT chassis volts | grep -A 2 Index |grep -v Index | grep -v "\-\-" | cut -c 29- > ${XYMONTMP}/xymon-hardware_volts.tmp while read LINE ; do echo $LINE | grep -q Status | grep -q Ok if [ $ERROR ] ; then PROBE_IN_ERROR="$LINE" echo "&yellow Le voltage de $PROBE_IN_ERROR est incorrect !" >> $MSG_FILE fi unset ERROR if [ $? -ne 0 ] ; then VOLT_YELLOW=1 ERROR=1 fi done < ${XYMONTMP}/xymon-hardware_volts.tmp fi if [ $VOLT_YELLOW ] ; then YELLOW=1 VOLT_GLOBAL_STATUS=yellow fi #Test ventilateurs : $OMREPORT chassis fans | grep -q "Main System Chassis Fans: Ok" if [ $? -eq 0 ] ; then FANS_GLOBAL_STATUS=green else $OMREPORT chassis fans | grep -A 6 Index |grep -v Index | grep -v "\-\-" |grep -v "N\/A" | cut -c 29- > ${XYMONTMP}/xymon-hardware_fans.tmp while read LINE ; do if [ $NEXT_LINE == FAN_MIN_RPM ] ; then FAN_MIN_RPM=$(echo $LINE | awk '{print $1}') echo "&yellow Le ventilateur $FAN_NAME tourne trop lentement ($FAN_RPM inferieur a ${FAN_MIN_RPM}) !!! ${FAN_NAME}_rpm: $FAN_RPM" >> ${XYMONTMP}/xymon-hardware_fans.msg unset NEXT_LINE fi if [ $NEXT_LINE == FAN_RPM ] ; then FAN_RPM=$(echo $LINE | awk '{print $1}') NEXT_LINE=FAN_MIN_RPM fi if [ $ERROR ] && [ $NEXT_LINE == FAN_NAME ] ; then FAN_IN_ERROR=$LINE NEXT_LINE=FAN_RPM if [ $FAN_RPM -le 0 ] ; then FAN_RED=1 echo "&red Le ventilateur $FAN_NAME ne tourne plus !!!" >> ${XYMONTMP}/xymon-hardware_fans.msg fi unset ERROR fi echo $LINE | grep -q Status | grep -q Ok if [ $? -ne 0 ] ; then FAN_YELLOW=1 ERROR=1 NEXT_LINE=FAN_NAME fi done < ${XYMONTMP}/xymon-hardware_fans.tmp fi if [ $FAN_RED ] ; then RED=1 echo "&red Probleme avec les vitesses des ventilateurs ! $(cat ${XYMONTMP}/xymon-hardware_fans.msg)" >> $MSG_FILE elif [ $FAN_YELLOW ] ; then YELLOW=1 echo "&yellow Probleme avec les vitesses des ventilateurs ! $(cat ${XYMONTMP}/xymon-hardware_fans.msg)" >> $MSG_FILE else VOLT_GLOBAL_STATUS=green echo "&green Tout va bien avec les ventilateurs" >> $MSG_FILE fi #Test etat disques : $OMREPORT storage pdisk controller=0 |grep ^Status | grep -q Ok if [ $? -eq 0 ] ; then echo "&green Le statut des disques est Ok !" >> $MSG_FILE else DISK_COLOR=yellow $OMREPORT storage pdisk controller=0 |grep -A 1 ^Status | grep -v "\-\-" > ${XYMONTMP}/xymon-hardware_disks.tmp while read LINE ; do echo $LINE | grep -q Status | grep -q Ok if [ $NEXT_LINE == DISK_NAME ] ; then DISK_NAME=$(echo $LINE | cut -c 29-) echo "&yellow Le disque $DISK_NAME est en mauvaise situation !" >> $MSG_FILE unset NEXT_LINE fi unset ERROR if [ $? -ne 0 ] ; then ERROR=1 NEXT_LINE=DISK_NAME fi done < ${XYMONTMP}/xymon-hardware_disks.tmp fi } function use_hpacucli () { $HPACUCLI ctrl all show config | grep drive | while read OUTPUT ; do TYPE=$(echo $OUTPUT | awk '{print $1}' | sed s/drive//) SLOT=$(echo $OUTPUT | awk '{print $2}') STATUS=$(echo $OUTPUT | awk '{print $NF}' | sed s/\)//) if [ $TYPE == "logical" ] ; then RAID=$(echo $OUTPUT | awk '{print $6}') SIZE=$(echo $OUTPUT | awk '{print $3 $4}' | sed s/\(// | sed s/\,//) if [ "$STATUS" != "OK" ] ; then RED=1 LINE="&red Logical drive $SLOT \(RAID $RAID, size : $SIZE\) status is BAD !!!" elif [ "$STATUS" == "OK" ] ; then LINE="&green Logical drive $SLOT \(RAID $RAID, size : $SIZE\) status is OK" else RED=1 LINE="&red Unknow status \(or stupid monitoring script\) for logical drive $SLOT \(RAID $RAID, size : $SIZE\) !!!" fi elif [ "$TYPE" == "physical" ] ; then SIZE=$(echo $OUTPUT | awk '{print $8 $9}' | sed s/\,//) if [ "$STATUS" != "OK" ] ; then YELLOW=1 LINE="&yellow Physical drive in slot $SLOT \(size : $SIZE\) status is BAD !!!" elif [ "$STATUS" == "OK" ] ; then LINE="&green Physical drive in slot $SLOT \(size : $SIZE\) status is OK" else RED=1 LINE="&red Unknow status \(or stupid monitoring script\) for physical drive in slot $SLOT \(size : $SIZE\) !!!" fi fi echo $LINE >> $MSG_FILE done } $GREP -q ^HPACUCLI=1 $CONFIG_FILE if [ $? -eq 0 ] ; then use_hpacucli fi $GREP -q ^SMARTCTL=1 $CONFIG_FILE if [ $? -eq 0 ] ; then use_smartctl fi $GREP -q ^HDDTEMP=1 $CONFIG_FILE if [ $? -eq 0 ] ; then use_hddtemp fi $GREP -q ^OPENMANAGE=1 $CONFIG_FILE if [ $? -eq 0 ] ; then use_openmanage fi $GREP -q ^SENSOR=1 $CONFIG_FILE if [ $? -eq 0 ] ; then use_lmsensors fi if [ "$RED" ] ; then FINAL_STATUS=red elif [ "$YELLOW" ] ; then FINAL_STATUS=yellow else FINAL_STATUS=green fi "$BB" "$BBDISP" "status "$MACHINE"."$TEST" "$FINAL_STATUS" $("$DATE") $("$CAT" "$MSG_FILE") "
Known Bugs and Issues
None
To Do
v0.6
- To be independent of /etc/sensors.conf → we get raw values, and we set right ones from those, and define thresolds in hobbit-hardware.conf file
- Support for independant temperatures thresolds for each disk
- Support for multiples sensors
- Support for multiples disk controllers chipsets
Credits
- Thanks to Xavier Carol i Rosell to remember me to upload new version, and for a fix proposal
Changelog
- 2009-01-15 v0.1
- Initial release.
- 2009-06-18 v0.1.1
- Bug fixes.
- 2009-06-25 v0.1.2
- New error messages (more verbose, more accurate).
- 2009-11-14 v0.2
- More verbosity when commands fail
- Disk temperature thresolds in hobbit-hardware.conf file.
- Support smartctl to replace hddtemp (if needed).
- Possibility to disable lm-sensors.
- Possibility to choose smartctl.
- 2010-01-22 v0.2.1
- Minor bug fix.
- 2013-06-27 v0.2.2
- Minor code optimizations
- 2011-09-09 v0.3
- Add support for OpenManage Physical disks, temps
- 2013-06-27 v0.4
- Fix hddtemp output handling (print last field instead of field N)
- 2013-09-27 v0.5
- Add support for HP monitoring tool (hpacucli)
- 2022-07-13 v0.6
- Add support for disks independent temperatures
</code>