Grafana Associate flashcards

32 cards. Read the term and the definition here. The deck at the top of this page is optional.

Grafana organization

A tenant-like grouping of users, teams, folders, and dashboards. A user can belong to several organizations and switch between them, and each organization has its own data sources and dashboards.

Grafana role hierarchy

Viewer can view dashboards; Editor can create and edit dashboards, folders, and panels; Admin can also manage users, teams, and data sources. Server Admin is a separate instance-wide role.

Explore view

An ad-hoc query workspace that runs queries without saving a panel. It shows raw results and a table view, making it the standard way to validate a data source or query.

Dashboard annotations

Point-in-time event markers drawn on panels. They can be added manually, sent through the annotations HTTP API, or produced by a data source query, and are scoped to a dashboard or organization.

Dashboard version history

A stored list of dashboard revisions with an optional message and author. Restoring an old revision creates a new version, so earlier revisions remain available in the history.

Provisioned dashboards

Dashboards loaded from files under provisioning/ or via the API with a provisioning flag. They are read-only in the UI, and edits must be made in the source files and reloaded.

Dashboard snapshot

An immutable, shareable copy of a dashboard including its rendered data. Later edits to the original dashboard do not change the snapshot, and the feature can be disabled.

Command palette shortcut

Ctrl/Cmd + K opens the command palette for jumping to dashboards, folders, and settings. Ctrl/Cmd + M toggles the left-hand navigation menu to free canvas space.

Data source UID

A stable identifier assigned to a data source. Dashboards and panels reference the UID, so renaming a data source does not break queries, but deleting and recreating it does.

$__rate_interval macro

A Grafana macro for Prometheus queries that resolves to a rate window aligned with the panel's scrape interval and time range, reducing empty or noisy rate() results.

$__interval macro

A Grafana macro replaced at render time with the time span between data points, derived from the panel width, Max data points, and Min interval settings.

LogQL stream selector

A Loki query starts with a label selector in braces, such as {job="api"}, optionally followed by line filters (|=, !=, |~) and parsers (json, logfmt) to extract fields.

SQL time macros

Grafana SQL macros such as $__timeFilter(column) and $__timeGroup(column, interval) inject the dashboard time range and bucket the results into aligned time groups.

Variable format modifiers

Suffixes on a variable reference, such as ${var:sqlstring}, ${var:regex}, ${var:csv}, or ${var:json}, that escape and format the value for the target query syntax.

Max data points setting

A per-query limit on returned points. Grafana uses it with panel width to compute $__interval, so lowering it produces coarser, faster queries and smoother graphs.

Mixed data source mode

A panel setting that lets each query row choose its own data source, so a single panel can combine Prometheus, Loki, and SQL results in one visualization.

Data source health check

The Save & test action on a data source settings page. It verifies credentials and connectivity and reports the backend error message when the check fails.

Time series panel

The default visualization for time-stamped data. It supports line, bar, and point styles plus axis units, thresholds, and per-field overrides for appearance control.

Field overrides

Per-field or per-series settings such as unit, color, thresholds, and decimals. Overrides apply after panel defaults and take precedence over them for matching fields.

Threshold modes

Absolute thresholds compare a value to fixed numbers; percentage thresholds compare to the field's min/max range. Thresholds drive color and can mark values as critical or warning.

Transformations

Operations applied after queries and before rendering, such as Join by field, Merge, Organize fields, Calculate field, Reduce, and Filter data by value. They reshape data without changing the source.

Stat and Gauge panels

Single-value visualizations that reduce a series using a calculation such as Last, Mean, or Max, controlled by the Reduce options in the panel editor.

Panel data links

Clickable links attached to values or fields. They can interpolate ${__value.text}, ${__field.name}, and ${__series.name} into the target URL for drill-down navigation.

Panel inspector

A panel-side debug view showing raw data, query statistics such as row count and query time, and the panel JSON. It is the fastest way to diagnose an empty panel.

Unit and decimals settings

Field options that control number formatting. Unit families such as bytes, seconds, percent, and currency drive axis labels, legend values, and threshold display.

Template variable types

Dashboard variables can be query, custom, textbox, constant, interval, data source, or ad hoc. Query variables populate options from a data source; custom variables use a fixed list.

Variable refresh setting

Controls when a variable reloads its options: on dashboard load, on time range change, or never. Dependent variables usually need refresh on time range change.

Chained variables

A variable whose query references an earlier variable, such as $env in a label filter. When the parent changes, Grafana re-runs the dependent query and updates its options.

Share link with time range

The Share dialog can lock the current time range and variable values into the URL, so the recipient sees exactly the same view when opening the link.

Dashboard JSON export

The Export or Save as JSON action produces the editable dashboard definition, including panels, variables, and queries, for version control or import into another instance.

Embedding requirement

Iframe embedding requires allow_embedding = true in the Grafana configuration. Without it, the browser blocks the iframe and the dashboard does not render.

Collapsed rows

Row panels group related panels and can be collapsed. Collapsed rows do not execute their queries until expanded, which reduces load on large dashboards.