Supported items

We created some items that use the Zabbix agent in both directions and gather data. But those are hardly the only ones available. You could check out the list while creating an item again (go to Configuration | Hosts, click on Items for any host, and click on the Create item button, followed by the Select button next to the Key field) in order to see which items are built in for Zabbix agents, along with a short description for most of them.

Not all Zabbix agent items are available as both passive and active items. For example, log and event log items (for gathering log file and Windows event log information, respectively) are only available as active items. Log monitoring is covered in Chapter 10, Advanced Item Monitoring, and Windows-specific items in Chapter 22, Monitoring Windows.

Looking at the list, we can find out which categories of items Zabbix agents support natively—system configuration, network traffic, network services, system load and memory usage, filesystem monitoring, and others. But that does not mean everything you see there will work on any system that the Zabbix agent daemon runs on. As every platform has a different way of exposing this information and some parameters might even be platform-specific, it isn't guaranteed that every key will work on every host.

For example, when the disk drive statistics report changes to userspace, the Zabbix agent has to specifically implement support for the new method; hence, older agent versions will support fewer parameters on recent Linux systems. If you are curious about whether a specific parameter works on a specific version of a specific operating system, the best way to find out is to check the Zabbix manual and then test it. Some of the most common agent item keys are as follows:

  • agent.ping: This returns 1 when the agent is available and nothing at all when the agent is not available
  • net.if.in/out/total: This provides incoming/outgoing or total traffic information
  • net.tcp.service: This tries to make a simplistic connection to a TCP service
  • proc.num: This counts the number of processes and can filter by various parameters
  • vfs.fs.size: This provides filesystem usage information
  • vm.memory.size: This provides memory usage information
  • system.cpu.load: This provides CPU load information in a standard decimal representation
  • system.cpu.util: This provides CPU utilization information, for example, iowait

For most of these, various parameters an be specified to filter the result or choose a particular piece of information. For example, proc.num[,zabbix] will count all processes that the Zabbix user is running.

..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset
3.16.29.209