Table des matières
- 2026:
- 2025:
12 billet(s) pour novembre 2026
| Notes conversion JSON YAML | 2026/11/13 11:16 | Jean-Baptiste |
| Git commit perdu | 2026/11/12 18:16 | Jean-Baptiste |
| Notes mojo lang | 2026/11/11 20:34 | Jean-Baptiste |
| Notes rust lang compilation | 2026/11/11 18:35 | Jean-Baptiste |
| Mémoire VSIZE VMSIZE SEGV lang C | 2026/11/11 15:59 | Jean-Baptiste |
| Ansible-navigator utilisation de ansible-vault | 2026/11/06 11:33 | Jean-Baptiste |
| Err Ansible - Module result deserialization failed No start of json char found | 2026/11/06 09:09 | Jean-Baptiste |
| Script Python conversion JSON vers CSV | 2026/11/06 08:14 | Jean-Baptiste |
| Filter Incoming IPv4 Packets Containing IP Options - cve-2026-98432 mitigation - Kernel TCP Use-After-Free | 2026/11/04 21:04 | Jean-Baptiste |
| Notes ZFS | 2026/11/03 18:04 | Jean-Baptiste |
| Diff image container | 2026/11/03 16:30 | Jean-Baptiste |
| Notes processus multi-thread | 2026/11/03 10:59 | Jean-Baptiste |
Support Ubuntu ESM Extended Security Maintenance
Voir :
sudo ua status
# dpkg -S /usr/bin/ua ubuntu-advantage-tools: /usr/bin/ua
Supervision ajout d'un hôte automatiquement dans Zabbix
Zabbix Auto Discovery (for active and for passive check)
Voir :
Le port TCP 10051 du serveur Zabbix doit-être joignable pour que les machines puissent si faire connaître de lui
Commenter la ligne Hostname=Zabbix server (pour que HostnameItem puisse s’appliquer)
/etc/zabbix/zabbix_agentd.conf
#Hostname= HostnameItem=system.hostname ServerActive=192.168.1.100 HostMetadataItem=system.uname
Redémarrer l'agent
service zabbix-agent restart
On jette un œil dans les logs
# tail -f /var/log/zabbix-agent/zabbix_agentd.log Configuration / Actions / Event source='Auto registration' Create action 'Conditions' tab : Host metadata like Linux 'Operations' tab : Add host Add to host groups: Virtual machines Link to templates: Template OS Linux
Via l'Api
Voir aussi :
Exemple de suppression
pip3 install --user zabbix-client
import zabbix_client s=zabbix_client.ZabbixServerProxy('https://acme.fr/zabbix') s.apiinfo.version() s.user.login(user='Admin', password='zabbix') s.host.get(output=['hostid', 'host']) # Ne plus superviser s.host.update({'hostid': '10124', 'status': '1'}) # Superviser de nouveau s.host.update({'hostid': '10124', 'status': '0'}) s.host.delete(10119)
Supervision - sonde Nagios - Mémoire Linux et CPU
Voir :
Voir SNMP :
Voir aussi :
Sonde Nagios supervision de la RAM & CPU via SNMPv3
Voir :
Voir aussi :
check_snmp_mem_cpu.sh
#! /bin/bash # Creative Commons CC0 Public Domain Licence set -euo pipefail IFS=$' \t\n' export LC_ALL=C SCRIPT_NAME=$(basename "$0") SNMP_HOST=localhost # DEFAULT #SNMP_PASSWD='P@ssw0rd' # YOU SHOULD USE ~/.snmp/snmp.conf FOR CREDS INSTEAD OF ARGUMENT PARAM #SNMP_USER=nagios # Pour iowait declare -i INTERVAL INTERVAL=5 exec 6>&1 # Link file descriptor #6 with stdout. Saves stdout. # FIX IGNORED ~/.snmp/snmp.conf WHEN 'sudo -u nagios check_snmp_host.sh' HOME=$(getent passwd "$(whoami)" | cut -d: -f6) # DEFAULT VALUES DEFAULT_MEM_WARNING_THRESHOLD=80 DEFAULT_MEM_CRITICAL_THRESHOLD=90 DEFAULT_DISK_WARNING_THRESHOLD=80 DEFAULT_DISK_CRITICAL_THRESHOLD=90 DEFAULT_CPU_WARNING_THRESHOLD=95 DEFAULT_CPU_CRITICAL_THRESHOLD=99 DEFAULT_SWAP_WARNING_THRESHOLD=80 DEFAULT_SWAP_CRITICAL_THRESHOLD=90 DEFAULT_IOWAIT_WARNING_THRESHOLD=20 DEFAULT_IOWAIT_CRITICAL_THRESHOLD=30 DEFAULT_USERS_WARNING_THRESHOLD=100 DEFAULT_USERS_CRITICAL_THRESHOLD=200 DEFAULT_PROCS_WARNING_THRESHOLD=600 DEFAULT_PROCS_CRITICAL_THRESHOLD=1000 DEFAULT_LOAD_WARNING_THRESHOLD=100,100,100 DEFAULT_LOAD_CRITICAL_THRESHOLD=200,200,200 PART_EXCLUDE=".*cdrom.*" OID_STORAGE_DESC=.1.3.6.1.2.1.25.2.3.1.3 OID_STORAGE_SIZE=.1.3.6.1.2.1.25.2.3.1.5 OID_STORAGE_USED=.1.3.6.1.2.1.25.2.3.1.6 OID_CPU_PERCENT_USED=.1.3.6.1.2.1.25.3.3.1.2 OID_SWAP_TOTAL=.1.3.6.1.4.1.2021.4.3.0 OID_SWAP_FREE=.1.3.6.1.4.1.2021.4.4.0 #OID_SWAP_OUT=.1.3.6.1.4.1.2021.11.4 OID_LOADAVG=1.3.6.1.4.1.2021.10.1.5 OID_CPU_IOWAIT=.1.3.6.1.4.1.2021.11.54.0 OID_USERS_COUNT=.1.3.6.1.2.1.25.1.5.0 OID_PROCS_COUNT=.1.3.6.1.2.1.25.4.2.1.2 exit_malfunction() { exec 1>&6 6>&- # Restore stdout and close file descriptor #6 echo "${LABEL:-UNDEF} UNKNOWN - ERROR $*" exit 3 } # NAGIOS EXIT STATUS # STATE_OK=0 ; STATE_WARNING=1 ; STATE_CRITICAL=2 ; STATE_UNKNOWN=3; STATE_DEPENDENT=4 STATUS_MSG=UNKNOWN typeset -i STATUS_CODE=3 trap 'if [ $STATUS_CODE -eq 3 ] && [ $? -ne 0 ] ; then exit_malfunction "Unexpected error" fi' EXIT ERR command_exists() { command -v "$1" >/dev/null 2>&1 || ( exit_malfunction "I require $1 but it's not installed. Aborting." ) } command_exists snmpget command_exists snmpwalk calc() { awk "BEGIN { print $* }"; } get_nproc() { snmpwalk -r 2 -v 3 "$SNMP_HOST" "$OID_CPU_PERCENT_USED" | wc -l || exit_malfunction "SNMP failed" } get_mem() { local -i USED local -i SIZE local -i i local -i MEM_PHYS_USED local -i MEM_PHYS_SIZE local -i MEM_BUFF_USED local -i MEM_CACH_USED local -i MEM_TOTA_USED local REAL_MEM_PHYS_USED_PERCENT for i in {10..1}; do # DESC="$(snmpget -r 2 -v 3 -a MD5 -A "$SNMP_PASSWD" -l authNoPriv -u "$SNMP_USER" "$SNMP_HOST" "${OID_STORAGE_DESC}.${i}" | cut -d= -f2 | cut -d: -f2)" || exit_malfunction DESC="$(snmpget -r 2 -v 3 "$SNMP_HOST" "${OID_STORAGE_DESC}.${i}" | cut -d= -f2 | cut -d: -f2)" || exit_malfunction "SNMP failed" if echo "$DESC" | grep -q -i -e "memory"; then USED="$(snmpget -r 2 -v 3 "$SNMP_HOST" "${OID_STORAGE_USED}.${i}" | cut -d= -f2 | cut -d: -f2)" || exit_malfunction "SNMP failed" SIZE="$(snmpget -r 2 -v 3 "$SNMP_HOST" "${OID_STORAGE_SIZE}.${i}" | cut -d= -f2 | cut -d: -f2)" || exit_malfunction "SNMP failed" fi if echo "$DESC" | grep -q -i -e "Physical memory"; then MEM_PHYS_USED="$USED" MEM_PHYS_SIZE="$SIZE" elif echo "$DESC" | grep -q -i -e "Memory buffers"; then MEM_BUFF_USED="$USED" elif echo "$DESC" | grep -q -i -e "Cached memory"; then MEM_CACH_USED="$USED" fi done # BECAUSE FREE MEMORY DONT MEAN AVAILABLE MEMORY MEM_TOTA_USED=$((MEM_PHYS_USED - MEM_BUFF_USED - MEM_CACH_USED)) REAL_MEM_PHYS_USED_PERCENT=$(calc $MEM_TOTA_USED \* 100/"$MEM_PHYS_SIZE") REAL_MEM_PHYS_USED_PERCENT=$(printf "%.2f" "$REAL_MEM_PHYS_USED_PERCENT") if [ "${REAL_MEM_PHYS_USED_PERCENT%.*}" -ge "${MEM_CRITICAL_THRESHOLD}" ]; then STATUS_MSG="CRITICAL" STATUS_CODE=2 elif [ "${REAL_MEM_PHYS_USED_PERCENT%.*}" -ge "${MEM_WARNING_THRESHOLD}" ]; then STATUS_MSG="WARNING" STATUS_CODE=1 else STATUS_MSG="OK" STATUS_CODE=0 fi echo "$LABEL $STATUS_MSG - ${REAL_MEM_PHYS_USED_PERCENT}% used ($((MEM_TOTA_USED / 1024)) MB / $((MEM_PHYS_SIZE / 1024)) MB)|TOTAL=$((MEM_PHYS_SIZE / 1024));USED=$((MEM_PHYS_USED / 1024));CACHE=$((MEM_CACH_USED / 1024));BUFFER=$((MEM_BUFF_USED / 1024))" } get_disk() { local -i PART_USED local -i PART_SIZE local LISTE local REGEX_PART local REGEX_EXCLUDE LISTE=$(snmpwalk -r 2 -v 3 "$SNMP_HOST" "$OID_STORAGE_DESC" | cut -d'=' -f1 | tr -d ' ' | grep -o '[0-9]\+$') || exit_malfunction "SNMP failed" if [ "${PART:-}" = "" ]; then REGEX_PART='^/' REGEX_EXCLUDE="^${PART_EXCLUDE}$" else REGEX_PART="^${PART%/}/?$" REGEX_EXCLUDE='^$' fi for i in $LISTE; do DESC="$(snmpget -r 2 -v 3 "$SNMP_HOST" "${OID_STORAGE_DESC}.${i}" | cut -d= -f2 | cut -d: -f2 | xargs)" || exit_malfunction "SNMP failed" if echo "$DESC" | grep -E -v -i -e "$REGEX_EXCLUDE" | grep -q -E -e "$REGEX_PART"; then PART_USED="$(snmpget -r 2 -v 3 "$SNMP_HOST" "${OID_STORAGE_USED}.${i}" | cut -d= -f2 | cut -d: -f2)" || exit_malfunction "SNMP failed" PART_SIZE="$(snmpget -r 2 -v 3 "$SNMP_HOST" "${OID_STORAGE_SIZE}.${i}" | cut -d= -f2 | cut -d: -f2)" || exit_malfunction "SNMP failed" PART_USED_PERCENT=$(calc $PART_USED \* 100/"$PART_SIZE") PART_USED_PERCENT=$(printf "%.2f" "$PART_USED_PERCENT") if [ "${PART_USED_PERCENT%.*}" -ge "${DISK_CRITICAL_THRESHOLD}" ]; then STATUS_MSG="CRITICAL" STATUS_CODE=2 elif [ "${PART_USED_PERCENT%.*}" -ge "${DISK_WARNING_THRESHOLD}" ]; then STATUS_MSG="WARNING" STATUS_CODE=1 else STATUS_MSG="OK" STATUS_CODE=0 fi if [ "${PART:-}" == "" ]; then if [ "$STATUS_CODE" -ne 0 ]; then echo "$LABEL $STATUS_MSG - DISK ${DESC} ${PART_USED_PERCENT}%|PART_USED=${PART_USED};PART_SIZE=${PART_SIZE}" break else echo "DEBUG: $LABEL $STATUS_MSG - DISK ${DESC} ${PART_USED_PERCENT}%|PART_USED=${PART_USED};PART_SIZE=${PART_SIZE}" >&2 fi else echo "$LABEL $STATUS_MSG - DISK ${DESC} ${PART_USED_PERCENT}%|PART_USED=${PART_USED};PART_SIZE=${PART_SIZE}" fi fi done if [ "${PART_USED_PERCENT:-}" = "" ]; then exit_malfunction "PART: ${PART:-} not found" fi if [ "${PART:-}" == "" ] && [ "$STATUS_CODE" -eq 0 ]; then echo "$LABEL $STATUS_MSG - DISK ALL OK" fi } get_cpu() { local -i MAX_CPU_PERCENT MAX_CPU_PERCENT="$(snmpwalk -r 2 -v 3 "$SNMP_HOST" "${OID_CPU_PERCENT_USED}" | cut -d= -f2 | cut -d: -f2 | sort -n | tail -1)" || exit_malfunction "SNMP failed" if [ "${MAX_CPU_PERCENT}" -ge "${CPU_CRITICAL_THRESHOLD}" ]; then STATUS_MSG="CRITICAL" STATUS_CODE=2 elif [ "${MAX_CPU_PERCENT}" -ge "${CPU_WARNING_THRESHOLD}" ]; then STATUS_MSG="WARNING" STATUS_CODE=1 else STATUS_MSG="OK" STATUS_CODE=0 fi echo "$LABEL $STATUS_MSG - MAX CORE USAGE: ${MAX_CPU_PERCENT}%|MAX_CORE_USAGE_PERCENT=${MAX_CPU_PERCENT}" } get_swap() { local -i SWAP_TOTAL local -i SWAP_FREE local SWAP_USED_PERCENT SWAP_TOTAL="$(snmpwalk -r 2 -v 3 "$SNMP_HOST" "${OID_SWAP_TOTAL}" | cut -d= -f2 | cut -d: -f2)" || exit_malfunction "SNMP failed" SWAP_FREE="$(snmpwalk -r 2 -v 3 "$SNMP_HOST" "${OID_SWAP_FREE}" | cut -d= -f2 | cut -d: -f2)" || exit_malfunction "SNMP failed" SWAP_USED=$((SWAP_TOTAL - SWAP_FREE)) SWAP_USED_PERCENT=$(calc $SWAP_USED \* 100/"$SWAP_TOTAL") SWAP_USED_PERCENT=$(printf "%.2f" "$SWAP_USED_PERCENT") if [ "${SWAP_USED_PERCENT%.*}" -ge "${SWAP_CRITICAL_THRESHOLD}" ]; then STATUS_MSG="CRITICAL" STATUS_CODE=2 elif [ "${SWAP_USED_PERCENT%.*}" -ge "${SWAP_WARNING_THRESHOLD}" ]; then STATUS_MSG="WARNING" STATUS_CODE=1 else STATUS_MSG="OK" STATUS_CODE=0 fi echo "$LABEL $STATUS_MSG - ${SWAP_USED_PERCENT}% used ( $((SWAP_USED / 1024)) MB / $((SWAP_TOTAL / 1024)) MB)|SWAP_TOTAL=${SWAP_TOTAL};SWAP_FREE=${SWAP_FREE}" } get_iowait() { declare -i IOWAIT_FIRST declare -i IOWAIT_SECOND IOWAIT_FIRST=$(snmpget -Oqv -v3 "$SNMP_HOST" "$OID_CPU_IOWAIT") || exit_malfunction "SNMP failed" sleep $INTERVAL IOWAIT_SECOND=$(snmpget -Oqv -v3 "$SNMP_HOST" "$OID_CPU_IOWAIT") || exit_malfunction "SNMP failed" IOWAIT_PERCENT=$(calc $((IOWAIT_SECOND - IOWAIT_FIRST)) / "$(get_nproc)" / $INTERVAL) IOWAIT_PERCENT=$(printf "%.2f" "$IOWAIT_PERCENT") if [ "${IOWAIT_PERCENT%.*}" -ge "${IOWAIT_CRITICAL_THRESHOLD}" ]; then STATUS_MSG="CRITICAL" STATUS_CODE=2 elif [ "${IOWAIT_PERCENT%.*}" -ge "${IOWAIT_WARNING_THRESHOLD}" ]; then STATUS_MSG="WARNING" STATUS_CODE=1 else STATUS_MSG="OK" STATUS_CODE=0 fi echo "$LABEL $STATUS_MSG - ${IOWAIT_PERCENT}%|IOWAIT_PERCENT=${IOWAIT_PERCENT}" } get_users_count() { declare -i USERS_COUNT USERS_COUNT="$(snmpget -r 2 -v 3 "$SNMP_HOST" "${OID_USERS_COUNT}" | cut -d= -f2 | cut -d: -f2)" || exit_malfunction "SNMP failed" if [ "${USERS_COUNT}" -ge "${USERS_CRITICAL_THRESHOLD}" ]; then STATUS_MSG="CRITICAL" STATUS_CODE=2 elif [ "${USERS_COUNT}" -ge "${USERS_WARNING_THRESHOLD}" ]; then STATUS_MSG="WARNING" STATUS_CODE=1 else STATUS_MSG="OK" STATUS_CODE=0 fi echo "$LABEL $STATUS_MSG - ${USERS_COUNT} users currently logged in|users=${USERS_COUNT}" } get_procs() { declare -i PROCS_COUNT PROCS_COUNT=$(snmpwalk -r 2 -v 3 "$SNMP_HOST" "${OID_PROCS_COUNT}" | wc -l) || exit_malfunction "SNMP failed" if [ "${PROCS_COUNT}" -ge "${PROCS_CRITICAL_THRESHOLD}" ]; then STATUS_MSG="CRITICAL" STATUS_CODE=2 elif [ "${PROCS_COUNT}" -ge "${PROCS_WARNING_THRESHOLD}" ]; then STATUS_MSG="WARNING" STATUS_CODE=1 else STATUS_MSG="OK" STATUS_CODE=0 fi echo "$LABEL $STATUS_MSG - ${PROCS_COUNT} processes|processes=${PROCS_COUNT}" } get_load() { local LOADAVG_1M_PERCENT local LOADAVG_5M_PERCENT local LOADAVG_15M_PERCENT LOADAVG_1M_PERCENT=$(snmpget -r 2 -v 3 "$SNMP_HOST" "${OID_LOADAVG}.1" | cut -d= -f2 | cut -d: -f2) || exit_malfunction "SNMP failed" LOADAVG_5M_PERCENT=$(snmpget -r 2 -v 3 "$SNMP_HOST" "${OID_LOADAVG}.2" | cut -d= -f2 | cut -d: -f2) || exit_malfunction "SNMP failed" LOADAVG_15M_PERCENT=$(snmpget -r 2 -v 3 "$SNMP_HOST" "${OID_LOADAVG}.3" | cut -d= -f2 | cut -d: -f2) || exit_malfunction "SNMP failed" LOADAVG_1M_PERCENT=$(calc $((LOADAVG_1M_PERCENT / $(get_nproc)))) LOADAVG_5M_PERCENT=$(calc $((LOADAVG_5M_PERCENT / $(get_nproc)))) LOADAVG_15M_PERCENT=$(calc $((LOADAVG_15M_PERCENT / $(get_nproc)))) local -i LOAD_WARN_1M local -i LOAD_WARN_5M local -i LOAD_WARN_15M local -i LOAD_CRIT_1M local -i LOAD_CRIT_5M local -i LOAD_CRIT_15M IFS=',' read -r LOAD_WARN_1M LOAD_WARN_5M LOAD_WARN_15M <<<"$LOAD_WARNING_THRESHOLD" IFS=',' read -r LOAD_CRIT_1M LOAD_CRIT_5M LOAD_CRIT_15M <<<"$LOAD_CRITICAL_THRESHOLD" if [ "${LOADAVG_1M_PERCENT%.*}" -ge "${LOAD_CRIT_1M}" ] || [ "${LOADAVG_5M_PERCENT%.*}" -ge "${LOAD_CRIT_5M}" ] || [ "${LOADAVG_15M_PERCENT%.*}" -ge "${LOAD_CRIT_15M}" ]; then STATUS_MSG="CRITICAL" STATUS_CODE=2 elif [ "${LOADAVG_1M_PERCENT%.*}" -ge "${LOAD_WARN_1M}" ] || [ "${LOADAVG_5M_PERCENT%.*}" -ge "${LOAD_WARN_5M}" ] || [ "${LOADAVG_15M_PERCENT%.*}" -ge "${LOAD_WARN_15M}" ]; then STATUS_MSG="WARNING" STATUS_CODE=1 else STATUS_MSG="OK" STATUS_CODE=0 fi echo "$LABEL $STATUS_MSG - LOAD AVERAGE: ${LOADAVG_1M_PERCENT}%, ${LOADAVG_5M_PERCENT}%, ${LOADAVG_15M_PERCENT}%|LOAD1=$LOADAVG_1M_PERCENT;LOAD5=$LOADAVG_5M_PERCENT;LOAD15=$LOADAVG_15M_PERCENT" } usage() { cat >&2 <<EOF Usage: $SCRIPT_NAME -t <mem|disk|cpu|swap|iowait|users|procs> [ -H <destination> ] [ -w <warnlevel> ] [ -c <critlevel> ] $SCRIPT_NAME -t disk [ -H <destination> ] [ -w <warnlevel> ] [ -c <critlevel> ] [ -p <mount_point> ] $SCRIPT_NAME -t disk [ -H <destination> ] [ -w <warnlevel> ] [ -c <critlevel> ] [ -e <exclude_mount_point_regex> ] Examples: $SCRIPT_NAME -t disk -p /var/lib $SCRIPT_NAME -t disk -e '/mnt/.*' $SCRIPT_NAME -t load -w 110,105,100 -c 160,130,100 EOF } main() { if [ "${TYPE:-}" = 'mem' ]; then LABEL=MEMORY typeset -i MEM_CRITICAL_THRESHOLD typeset -i MEM_WARNING_THRESHOLD MEM_CRITICAL_THRESHOLD=${CRITICAL_THRESHOLD:-$DEFAULT_MEM_CRITICAL_THRESHOLD} MEM_WARNING_THRESHOLD=${WARNING_THRESHOLD:-$DEFAULT_MEM_WARNING_THRESHOLD} get_mem elif [ "${TYPE:-}" = 'disk' ]; then LABEL=DISK typeset -i DISK_CRITICAL_THRESHOLD typeset -i DISK_WARNING_THRESHOLD DISK_CRITICAL_THRESHOLD=${CRITICAL_THRESHOLD:-$DEFAULT_DISK_CRITICAL_THRESHOLD} DISK_WARNING_THRESHOLD=${WARNING_THRESHOLD:-$DEFAULT_DISK_WARNING_THRESHOLD} get_disk elif [ "${TYPE:-}" = 'cpu' ]; then LABEL=CPU typeset -i CPU_CRITICAL_THRESHOLD typeset -i CPU_WARNING_THRESHOLD CPU_CRITICAL_THRESHOLD=${CRITICAL_THRESHOLD:-$DEFAULT_CPU_CRITICAL_THRESHOLD} CPU_WARNING_THRESHOLD=${WARNING_THRESHOLD:-$DEFAULT_CPU_WARNING_THRESHOLD} get_cpu elif [ "${TYPE:-}" = 'swap' ]; then LABEL=SWAP typeset -i SWAP_CRITICAL_THRESHOLD typeset -i SWAP_WARNING_THRESHOLD SWAP_CRITICAL_THRESHOLD=${CRITICAL_THRESHOLD:-$DEFAULT_SWAP_CRITICAL_THRESHOLD} SWAP_WARNING_THRESHOLD=${WARNING_THRESHOLD:-$DEFAULT_SWAP_WARNING_THRESHOLD} get_swap elif [ "${TYPE:-}" = 'iowait' ]; then LABEL=IOWAIT typeset -i IOWAIT_CRITICAL_THRESHOLD typeset -i IOWAIT_WARNING_THRESHOLD IOWAIT_CRITICAL_THRESHOLD=${CRITICAL_THRESHOLD:-$DEFAULT_IOWAIT_CRITICAL_THRESHOLD} IOWAIT_WARNING_THRESHOLD=${WARNING_THRESHOLD:-$DEFAULT_IOWAIT_WARNING_THRESHOLD} get_iowait elif [ "${TYPE:-}" = 'users' ]; then LABEL=USERS typeset -i USERS_CRITICAL_THRESHOLD typeset -i USERS_WARNING_THRESHOLD USERS_CRITICAL_THRESHOLD=${CRITICAL_THRESHOLD:-$DEFAULT_USERS_CRITICAL_THRESHOLD} USERS_WARNING_THRESHOLD=${WARNING_THRESHOLD:-$DEFAULT_USERS_WARNING_THRESHOLD} get_users_count elif [ "${TYPE:-}" = 'procs' ]; then LABEL=PROCS typeset -i PROCS_CRITICAL_THRESHOLD typeset -i PROCS_WARNING_THRESHOLD PROCS_CRITICAL_THRESHOLD=${CRITICAL_THRESHOLD:-$DEFAULT_PROCS_CRITICAL_THRESHOLD} PROCS_WARNING_THRESHOLD=${WARNING_THRESHOLD:-$DEFAULT_PROCS_WARNING_THRESHOLD} get_procs elif [ "${TYPE:-}" = 'load' ]; then LABEL=LOAD LOAD_CRITICAL_THRESHOLD=${CRITICAL_THRESHOLD:-$DEFAULT_LOAD_CRITICAL_THRESHOLD} LOAD_WARNING_THRESHOLD=${WARNING_THRESHOLD:-$DEFAULT_LOAD_WARNING_THRESHOLD} get_load else LABEL=UNDEF usage exit 3 fi } while [ "${1-}" != "" ]; do case $1 in -h | --help) usage exit 0 ;; -t | --type) shift TYPE=$1 ;; -c | --critical) shift CRITICAL_THRESHOLD=$1 ;; -w | --warning) shift WARNING_THRESHOLD=$1 ;; -H | --hostname) shift SNMP_HOST=$1 ;; -p | --partition) shift PART=$1 ;; -e | --exclude-partition) shift PART_EXCLUDE=$1 ;; --) # End of all options shift break ;; -*) echo "$SCRIPT_NAME: invalid option" >&2 echo "Try '$SCRIPT_NAME --help' for more information." >&2 exit 1 ;; *) usage ;; esac shift done main exit $STATUS_CODE
Conf SNMPv3 pour tester ce script :
/etc/snmp/snmpd.conf
skipNFSInHostResources 1 # rouser: a SNMPv3 read-write user # arguments: user [noauth|auth|priv] [restriction_oid] rouser nagios auth #rouser nagios
/usr/share/snmp/snmpd.conf (ou /var/net-snmp/snmpd.conf)
# DO NOT STORE CONFIGURATION ENTRIES HERE. # Please save normal configuration tokens for snmpd in SNMPCONFPATH/snmpd.conf. # Only "createUser" tokens should be placed here by snmpd administrators. # rouser nagios auth createUser nagios MD5 P@ssw0rd
Pb Timeout snmpget
On peut avoir des problèmes de Timeout SNMP. Pour pallier à ce problème il peut être utile de mettre skipNFSInHostResources 1 dans /etc/snmp/snmpd.conf
Sonde Nagios supervision de la RAM en local via la commande free
Source : https://exchange.nagios.org/directory/Plugins/System-Metrics/Memory/check_mem_ng-2Esh/details
check_mem_ng.sh
#!/bin/bash ################################################################################ #This plugin is loosely inspired by check_mem v1.1 from Lukasz Gogolin #https://exchange.nagios.org/directory/Plugins/System-Metrics/Memory/check_mem-2Esh/details #I did a code cleanup and added a check of "free" version, because since #procps-ng 3.3.0 (RHEL 7+) check_mem.sh doesn't work anymore (buff/cache column) ################################################################################ # 1.1 - Added -l flag and LEGACY_PERFATA var to put legacy perfdata mode ################################################################################ #Nagios Constants STATE_OK=0 STATE_WARNING=1 STATE_CRITICAL=2 STATE_UNKNOWN=3 SCRIPTPATH=`echo $0 | /bin/sed -e 's,[\\/][^\\/][^\\/]*$,,'` if [[ -f ${SCRIPTPATH}/utils.sh ]]; then . ${SCRIPTPATH}/utils.sh # use nagios utils to set real STATE_* return values fi #Useful functions printversion(){ echo "$0 $VERSION" echo } printusage() { printversion echo "Usage:" echo " check_mem_ng.sh [-w <warnlevel>] [-c <critlevel>] [-v] [-l]" echo " checks local host available memory" echo " warnlevel and critlevel is percentage value without %" echo " defaults being respectively 80 et 90" echo " add -v for verbose (debuging purpose)" echo " add -l for legacy perfdata mode (or change LEGACY_PERFDATA variable in script)" echo " check_mem_ng.sh -V" echo " prints version" echo " check_mem_ng.sh -h" echo " prints help (this message)" } printvariables() { echo "Variables:" #Add all your variables at the en of the "for" line to display them in verbose for i in WARNING_THRESHOLD CRITICAL_THRESHOLD FINAL_STATE FINAL_COMMENT LEGACY_PERFDATA FREE_OUTPUT TOTAL_MEM FREE_MEM BUFFCACHE_MEM BUFF_MEM CACHE_MEM USED_MEM TOTAL_MEM_MB USED_MEM_MB WARNING_THRESHOLD_B CRITICAL_THRESHOLD_B USED_MEM_PRC ENABLE_PERFDATA VERSION do echo -n "$i : " eval echo \$${i} done echo } #Set to unknown in case of unplaned exit FINAL_STATE=$STATE_UNKNOWN FINAL_COMMENT="UNKNOWN: Unplaned exit. You should check that everything is alright" #Default values WARNING_THRESHOLD=80 CRITICAL_THRESHOLD=90 ENABLE_PERFDATA=1 VERSION="1.1" VERBOSE=0 #####FORCE LEGACY MODE##### #put 1 to force legacy perfdata mode without using "-l" flag (no configuration change in nrpe.cfg) LEGACY_PERFDATA=0 #####FORCE LEGACY MODE##### #Process arguments while getopts ":c:hlvVw:" opt; do case $opt in c) CRITICAL_THRESHOLD=$OPTARG ;; h) printusage exit $STATE_OK ;; l) LEGACY_PERFDATA=1 ;; v) echo "Verbose mode ON" echo VERBOSE=1 ;; V) printversion exit $STATE_UNKNOWN ;; w) WARNING_THRESHOLD=$OPTARG ;; \?) echo "UNKNOWN: Invalid option: -$OPTARG" exit $STATE_UNKNOWN ;; :) echo "UNKNOWN: Option -$OPTARG requires an argument." exit $STATE_UNKNOWN ;; esac done #Real check begins here FREE_OUTPUT=`free -b | grep Mem:` TOTAL_MEM=`echo $FREE_OUTPUT |awk '{print $2}'` FREE_MEM=`echo $FREE_OUTPUT |awk '{print $4}'` if [ `free -V | grep procps-ng | wc -l` -eq 1 ]; then #procps-ng, free will display buff/cache as one column BUFFCACHE_MEM=`echo $FREE_OUTPUT |awk '{print $6}'` if [ $LEGACY_PERFDATA -eq 1 ] ; then #Get BUFFER and CACHE separatly for legacy output FREE_OUTPUT2=`free -b | grep Mem:` BUFF_MEM=`echo $FREE_OUTPUT2 |awk '{print $6}'` CACHE_MEM=`echo $FREE_OUTPUT2 |awk '{print $7}'` fi else #procps, free will display buff/cache as two separate columns BUFF_MEM=`echo $FREE_OUTPUT |awk '{print $6}'` CACHE_MEM=`echo $FREE_OUTPUT |awk '{print $7}'` BUFFCACHE_MEM=$(($BUFF_MEM+$CACHE_MEM)) fi #Deduce used memory from total/free/buffer+cache USED_MEM=$(($TOTAL_MEM-$FREE_MEM-$BUFFCACHE_MEM)) #Convert in MB for pseudo "human readable" plugin output. Todo, real human readable? TOTAL_MEM_MB=`echo "$TOTAL_MEM / 1048576" | bc` USED_MEM_MB=`echo "$USED_MEM / 1048576" | bc` WARNING_THRESHOLD_B=`echo "$WARNING_THRESHOLD * $TOTAL_MEM / 100" | bc` CRITICAL_THRESHOLD_B=`echo "$CRITICAL_THRESHOLD * $TOTAL_MEM / 100" | bc` #Convert in percents used memory USED_MEM_PRC=$((($USED_MEM*100)/$TOTAL_MEM)) #Check if available memory is below thresholds if [ "$USED_MEM_PRC" -ge "$CRITICAL_THRESHOLD" ]; then FINAL_STATE=$STATE_CRITICAL FINAL_COMMENT="CRITICAL: Memory above critical threshold. Total: ${TOTAL_MEM_MB} MB - Used: ${USED_MEM_MB} MB - ${USED_MEM_PRC}% used!" elif [ "$USED_MEM_PRC" -ge "$WARNING_THRESHOLD" ]; then FINAL_STATE=$STATE_WARNING FINAL_COMMENT="WARNING: Memory above warning threshold. Total: ${TOTAL_MEM_MB} MB - Used: ${USED_MEM_MB} MB - ${USED_MEM_PRC}% used!" else FINAL_STATE=$STATE_OK FINAL_COMMENT="OK: Memory below thresholds. Total: ${TOTAL_MEM_MB} MB - Used: ${USED_MEM_MB} MB - ${USED_MEM_PRC}% used." fi #Perfdata processing if [ $ENABLE_PERFDATA -eq 1 ] ; then #Added check_mem.sh legacy perfdata mode, to ease migration if [ $LEGACY_PERFDATA -eq 1 ] ; then PERFDATA=" | TOTAL=$TOTAL_MEM;;;; USED=$USED_MEM;;;; CACHE=$CACHE_MEM;;;; BUFFER=$BUFF_MEM;;;;" else PERFDATA=" | Memory_Used=${USED_MEM}B;$WARNING_THRESHOLD_B;$CRITICAL_THRESHOLD_B;0;$TOTAL_MEM;" fi fi #Script end, display verbose information if [[ $VERBOSE -eq 1 ]] ; then printvariables fi echo ${FINAL_COMMENT}${PERFDATA} exit $FINAL_STATE
Suivre la conso CPU par utilisateurs
Voir aussi :
- atop
Source :
GNU Accounting Utilities est un ensemble d'utilitaires qui rapportent et résument les données de statistiques sur les temps de connexion des utilisateurs et d'exécution de processus
La comptabilisation de connexions (« login accounting ») fournit un résumé sur l'utilisation des ressources système basé sur le temps de connexion et la comptabilisation de processus (« process accounting ») fournit un résumé basé sur les commandes exécutées sur le système.
Install RedHat
dnf install psacct # accton on systemctl enable --now psacct.service # Tester ls # Une commande bidon lastcomm # Doit retourner quelque chose
Install Debian
apt-get install acct # TODO: à compléter
Temps de connexion global d'un utilisateur (en heures)
ac
Temps de connexion global par utilisateur
ac -p
Temps de connexion global par jours
ac -d
Temps de connexion global par jours et par utilisateur
ac -p -d
Temps de connexion de l'utilisateur root
ac root
Temps de connexion de l'utilisateur root par jour
ac -d root
Commande sa
Affiche l'activité par utilisateur
sa -u
Affiche un résumé des ressources utilisées par utilisateurs
sa -m
Utilisation classé par pourcentage
sa -c
Liste les dernières commandes éxécutées
lastcomm
Dernière commandes executées pour l'utilisateur roger
lastcomm roger
Diagnostic
journalctl -u acct -u psacct --since "10 minutes ago" --no-pager
Exemple commande sa - mémoire Linux
# sa -u --other-acct-file /var/account/pacct-20260917 sa -u sa -m sa -p sa -P
# sa -u --other-acct-file /var/account/pacct-20260917 | sort -nk4 | tail -9 ansible 0.07 cpu 22400k mem 0 io systemd root 0.14 cpu 25104k mem 0 io vdo root 0.02 cpu 30736k mem 0 io gmain ansible 0.02 cpu 35248k mem 0 io sshd * ansible 0.00 cpu 38864k mem 0 io (sd-pam) * root 0.00 cpu 43920k mem 0 io (sd-sync) * root 0.93 cpu 121072k mem 0 io tuned root 0.00 cpu 370816k mem 0 io runc polkitd 0.05 cpu 435840k mem 0 io JS Helper
# sa -m --other-acct-file /var/account/pacct-20260917 | sort -nk6
421 337.66re 0.03cp 0avio 8654k
libstoragemgmt 1 6.21re 0.00cp 0avio 2154k
500 300 0.02re 0.00cp 0avio 5016k
root 107 297.62re 0.02cp 0avio 13654k
sshd 7 0.01re 0.00cp 0avio 16560k
ansible 5 27.71re 0.00cp 0avio 24716k
polkitd 1 6.08re 0.00cp 0avio 435840k
# sa -a --other-acct-file /var/account/pacct-20260917 | sort -nk5 | tail -4
1 0.00re 0.00cp 0avio 43920k (sd-sync)*
1 6.11re 0.02cp 0avio 121072k tuned
1 0.00re 0.00cp 0avio 370816k runc
1 6.08re 0.00cp 0avio 435840k JS Helper
# sa -c --other-acct-file /var/account/pacct-20260917 | head -4
421 100.00% 337.66re 100.00% 0.03cp 100.00% 0avio 8654k
29 6.89% 84.00re 24.88% 0.02cp 90.07% 0avio 44968k ***other
2 0.48% 10.61re 3.14% 0.00cp 4.64% 0avio 5371k bash
9 2.14% 12.14re 3.60% 0.00cp 1.99% 0avio 17284k sshd
# sa --sort-cpu-avmem --other-acct-file /var/account/pacct-20260917 | head -4
421 337.66re 0.03cp 0avio 8654k
29 84.00re 0.02cp 0avio 44968k ***other*
25 0.00re 0.00cp 0avio 19776k systemd-cgroups
8 5.70re 0.00cp 0avio 18896k sshd*
lastcomm --show-paging --pid
Err
Err - ac: Possible overflow of time_t! Can't continue.
Avec des logs en provenance d'une autre machine
$ ac -f pacct-20261012 -p pacct-20261012:3450: problem: update user (for logout) with -11 seconds?! ac: Possible overflow of time_t! Can't continue.
$ ac -f pacct-20261011 --complain 2>/dev/null | awk -F':' '!($1="") !($2="")' | sed -e 's/time warp .*/time warp /g' | sort -u problem empty record problem missing login record for `�' problem missing login record for `J' problem time warp problem update user (for logout) with -1 seconds?!
Se passer de Flash player
Pour les vidéo
(à l'époque beaucoup de vidéo nécessitaient Flash player)
Un solution ?
apt-get install browser-plugin-vlc
livestreamer :
- Dailymotion
- Livestream
- Twitch
- UStream
- YouTube Live
clive / cclive unplug Viewtube / Linterna Magica
rtmpdump remplacé par curl ?
yt-dlp (youtube-dl)
sudo apt-get install yt-dlp
youtube-dl http://rg3.github.io/youtube-dl/download.html
python3 -m pip install --user --upgrade youtube_dl
Pb yt-dlp
Err yt-dlp ERROR: unable to open for writing: [Errno 36] File name too long:
Solution
yt-dlp -o "%(id)s.%(ext)s" "https://twitter.com/JuliaDavisNews/status/1574860347697205262"
Pour les images vectorielles
- Gnash
