Documentation Plugins Zabbix Query editor

Zabbix query editor

The Zabbix query editor lets you build queries to visualize monitoring data from Zabbix. You can query numeric metrics, text data, problems, triggers, IT services, and user macros. Each query type has its own set of fields and options.

Before you begin

• Configure the Zabbix data source.
• Verify your Zabbix user has permissions to access the host groups and hosts you want to query.

Key concepts

If you’re new to Zabbix, the following terms are used throughout this documentation:

Query types

Select a query type from the drop-down at the top of the query editor. The available fields change based on the selected type.

Metrics

Use Metrics queries to retrieve numeric time series data from Zabbix items. This is the most common query type for building graphs and dashboards.

Expand the Options section to access additional settings:

Text

Use Text queries to retrieve text and log data from Zabbix items with text value types (character, log, text).

Expand the Options section to access additional settings:

Services

Use Services queries to retrieve IT service SLA data from Zabbix.

This query type shares the same Options as Metrics (Trends, Show disabled items, Use Zabbix value mapping, Disable data alignment).

Item ID

Use Item ID queries to retrieve data for specific Zabbix items by their numeric IDs. This is useful when you know the exact item IDs.

This query type shares the same Options as Metrics (Trends, Show disabled items, Use Zabbix value mapping, Disable data alignment).

Triggers

Use Triggers queries to count triggers matching specific criteria. This query type returns numeric data suitable for time series panels.

Expand the Options section to access additional settings:

Problems

Use Problems queries to retrieve Zabbix problem events. This query type returns tabular data suitable for table panels and the Problems panel.

Expand the Options section to access additional settings:

User macros

Use User macros queries to retrieve Zabbix user macro values.

Use regex in queries

You can use JavaScript regular expressions in the Group, Host, Application, Item tag, and Item fields to match multiple values. Wrap regex patterns in forward slashes (/pattern/).

Select multiple items

To display multiple CPU metrics on one graph, excluding idle time, use a regex in the Item field:

Copy

/CPU (?!idle).* time/

This matches items like “CPU user time”, “CPU system time”, and “CPU iowait time”, but excludes “CPU idle time”.

Compare metrics across hosts

To compare the same metric across multiple hosts, use regex in the Host field. For example, to show CPU system time for all hosts with names starting with “backend”:

• Group: /.*/
• Host: /^backend/
• Item: CPU system time

Match all values

Use /.*/ to match all values in a field. For example, setting Group to /.*/ queries across all host groups.

Apply functions

You can add processing functions to transform and aggregate query results when using Metrics, Item ID, or Services query types. Click the + button next to the query to add functions such as groupBy, scale, delta, rate, and movingAverage.

For a complete list of available functions, refer to the functions reference.

Direct DB Connection behavior

When Direct DB Connection is enabled, the plugin retrieves history and trend data directly from the Zabbix database instead of the Zabbix API. This is transparent to the query editor – you build queries the same way. The key difference is that the database performs server-side aggregation, which reduces data transfer and improves performance on wide time ranges.

The consolidateBy function directly controls the aggregation function used in database queries. When using Direct DB Connection, pair it with groupBy for accurate results. For example, to group values by 1-hour intervals using the maximum value:

Copy

consolidateBy(max) | groupBy(1h, max)

Without consolidateBy, the database aggregates using the default AVG function, which may produce unexpected results when combined with groupBy(interval, max).

Next steps

• Apply functions to transform query results
• Use template variables for dynamic dashboards
• Set up alerting rules