Skip to content
Branch: master
Clone or download
Marc Schoechlin
Marc Schoechlin Docuementation improvement
Latest commit 499e24d Feb 9, 2019
Type Name Latest commit message Commit time
Failed to load latest commit information.
zabbix_templates updated templates Dec 18, 2018
.travis.yml change release Feb 8, 2019



A set of zabbix UserParameter scripts and a monitoring template for linux systems.

Monitoring details

This package provides the following capabilities:

  • linux standard monitoring using the standard items provided by the zabbix-agent
    • monitor memory behavior
    • monitor important servcies: smtp, ssh, cron
    • swap usage
    • 5min system load
    • monitor dmesg for bad behavior of the system
    • monitor the maximum and minimum of processes
    • automatic discovery
      • filesystems: inode and space measures
        (the amount of discovered filesystems can be limited by a configuration file on the monitored host)
      • network interfaces: packets and transferrates per second
      • storage devices: operations per second (the amount of discovered devices can be limited by a configuration file on the monitored host)
      • network interfaces: packets and transferrates per second
    • number of processes
  • monitor ICMP ping
  • monitor the MTA mailqueue
  • monitor NFS operations/retransmits
  • apache monitoring: (enable /server-status for localhost clients,
    • Check Apache Server-Status (Readers, Writers - alert if to many slots are in use)
    • Loadbalancer check
    • Monitor Mod JK backend status
  • elasticsearch node and cluster monitoring (needs elasticsearch and es_stats_zabbix Python modules)
  • redis Monitoring (needs redis Python module)
  • NGINX Monitoring (enable /basic_status only for localhost clients,
  • generic discovery (put json snippets to /var/run/zabbix-generic-discovery/ITEMNAME.json i.e. with puppet and get a combined discovery value)
  • monitor puppet state (execution statistics, disabled state, ...)
  • linux software raid state
  • check quality of ntp sync behavior
  • check for required reboot
  • monitor open file descriptors
  • Configurable autodiscovery of block devices
  • Monitor performance havior of discovered block devices
  • Configurable autodisovery of real disk devices
  • Check smart state of disks and gather statistics
  • Discover network devices and monitor performance and error behavior
  • Zabbix agent version

Almost all measures are integrated in graphs which are displayed on the various host screens.

The project inlcudes various zabbix templates for relase 3.4.

Note: There are also 2.2 and 3.0 templates, but they are outdated because we do not support them anymore. Probably these outdated templates might be a good starting point for your own environment.

A quick is provided by the following files:

How to test and debug

  • Install the packages on the zabbix agent host

  • Test the agent on the agent machine

    zabbix_agentd  -t linux.dmesg
    zabbix_agentd --print
  • Test the agent on the server machine

    apt-get install zabbix-get
    zabbix_get -s -k linux.dmesg

How to release and/or install the userparameter scripts

Adpoted from:

  • Install FPM

    sudo apt-get install ruby-dev build-essential debhelper devscripts rpm
    gem install fpm --user
  • Get the repo

    git clone
    cd zabbix-agent-extensions
  • Release only: Edit this file ( and describe the new feature

  • Create packages

    git commit ....
  • Install the rpm or debian archive on as an addition to your zabbix-agent:

    rpm -Uvh noarch/zabbix-agent-extensions-<version>.noarch.rpm
    dpkg -i zabbix-agent-extensions_<version>_all.deb
  • Push release

    git tag <release-number>
    git push
    git push --tags
  • TravisCI builds and tests the release and uploads it to github

How to configure discovery for zabbix agent

  • Configure disk device discovery
    • Create files: (if the file does not exist, the default is used)
      • /etc/zabbix/item_zabbix_device_discovery.json
      • /etc/zabbix/item_zabbix_discovery_filesystems.json
    • Add content to include/exclude devices
      (what it does: include all devices and hardware models, after that filter out all devices and models which match to one of the python regexes)
      /usr/bin/zabbix_discovery_devices --help
      /usr/bin/zabbix_discovery_filesystems --help
    • Test
    zabbix_agentd -t ""
    /usr/bin/zabbix_discovery_devices --config /etc/zabbix/item_zabbix_discovery_devices.json --debug
    zabbix_agentd -t "vfs.fs.discovery"
    /usr/bin/zabbix_discovery_filesystems --config /etc/zabbix/item_zabbix_discovery_filesystems.json --debug
  • Configure generic discovery
    • Decide to use a appname consiting of the following characters: a-zA-Z0-9. (Alphanumeric and dot characters)
    • Add file snippes to /var/run/zabbix-generic-discovery/ which look like -.json
    • Test the disovery by
       zabbix_agentd -t "generic.discovery[appname]"

How to configure the zabbix server/templates

  • Open Zabbix web frontend
  • Open "Configuration" => "Templates"
  • Press button "Import"
  • Activate Linux template
    • Load "zabbix_templates/3.4/Custom - OS - Linux.xml"
    • Open template "Custom - OS - Linux" and modify the default values defined in the macros
      • Devices
        • {$DISK_HIGH_READ_IOPS_LIMIT} : alert limit for read iops/sec
        • {$DISK_HIGH_WRITE_IOPS_LIMIT} : alert limit for read iops/sec
        • {$DISK_IOPS_LIMIT_MEASURES} : how many measures need to be above the configured limits before alerting
      • Filesystems
        • {$DISK_USAGE_ABOVE_1TB_MINFREE_GBYTES} : how many gigagbytes should be free for filesystem above 1TB
        • {$DISK_USAGE_PERCENT_ALARM} : send alarm for filesystems below 1 TB on prcentage
        • {$DISK_USAGE_PERCENT_WARN} : send warning for filesystems below 1 TB on prcentage
      • {$MAXIMUM_NUMBER_RETRANSMISSIONS} : how many nfs retransmissions are ok on every measurement cycle
      • NTP
        • {$MAX_NTP_OFFSET_MS} : the maximum offset limit in milliseconds
        • {$MIN_NTP_SERVER_COUNT} : how many good ntp sources should be avaiilable, it is istrongly recommended to change the default to 2
      • {$MONITOR_LOAD_WARNING_MULT} : a multiplicator with the number of cpus for load monitoring
      • {$MONITOR_TIMEOUT} : amount of time to complain if hosts does not provide values anymore
    • Assign template "Custom - OS - Linux" to the desired hosts and modify the default values to host specific settings
  • Activate Apache template
    • Load "zabbix_templates/Custom - Service - Apache.xml"
    • Open template "Custom - Service - Apache" and modify the default values defined in the macros
    • Assign template "Custom - Service - Apache" to the desired hosts and modify the default values to host specific settings

Licence and Authors

Additional authors are very welcome - just submit your patches as pull requests.

This software is licensed by GPLv2 - review file "LICENSE"

You can’t perform that action at this time.
You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session.