Grafana

marmotdata/grafana

Discover dashboards, panels and data sources from Grafana

Dashboards Assets Lineage Experimental

The Grafana plugin discovers dashboards, their panels and the configured data sources from a Grafana instance. Panels backed by a SQL data source are linked to the tables their queries read.

Assets

Asset	Name	Notes
Dashboard	`<folder>/<title>`, or `<title>` in the General folder	Description, tags, version, time range and refresh interval from Grafana
Chart	`<dashboard name>/<panel title>`	One per panel; untitled panels are `panel-<id>`. Rows are flattened, text panels skipped. SQL, PromQL and LogQL queries are recorded on the chart
DataSource	The data source name	Type, address and database. Credentials are never read

Lineage

Dashboard CONTAINS Chart
DataSource FEEDS Chart, for every data source a panel or one of its queries uses
Table FEEDS Chart and Table FEEDS Dashboard, for panels querying a PostgreSQL, MySQL, SQL Server or ClickHouse data source. Tables are named the way the native Marmot plugin for that database names them, so the edges land on assets discovered by those plugins. Template variables and Grafana macros in a query never produce an edge.

Authentication

Create a service account in Grafana (Administration, Service accounts) and add a token to it. Tokens start with glsa_.

The Viewer role is enough: it reads dashboards, library panels and the data source list (names and addresses, never credentials). Should a token not be allowed to list data sources, the run still completes; charts then record the data source type and uid from the dashboard and no DataSource assets are created.

Configuration

host: "https://grafana.company.com"
api_key: "glsa_..."
include_panels: true
include_datasources: true
discover_lineage: true
page_size: 100
verify_ssl: true

In the UI

Point-and-click, no config file needed.

1 Open Runs Create pipeline
2 Pick Grafana from the plugin list.
3 Fill in the wizard, set a schedule, save.

Read the UI guide

With the CLI

Save a YAML config, then run marmot ingest.

ingest.yaml yaml

name: my-grafana-pipeline
runs:
  - grafana:
      host: "<host>"
      api_key: "***"

$ marmot ingest -c ingest.yaml

Read the CLI guide

Not using plugins? Other ways to populate Marmot

Terraform Declare assets as infrastructure-as-code. Pulumi Native TypeScript, Python, or Go resources. Kubernetes Operator Sync assets from Kubernetes custom resources.

Configuration

10 top-level fields. * marks required fields.

tags multiselect

Tags to apply to discovered assets

external_links object[]

External links to show on all assets

name string

Display name for the link

icon string

Icon identifier for the link

url string

URL to the external resource

filter object

Filter discovered assets by name (regex)

include multiselect

Include patterns for resource names (regex)

exclude multiselect

Exclude patterns for resource names (regex)

host string

Grafana URL, for example https://grafana.company.com

api_key password

Token of a Grafana service account

verify_ssl bool

Verify the TLS certificate of the Grafana host

default: true

include_panels bool

Discover dashboard panels as Chart assets

default: true

include_datasources bool

Discover data sources as DataSource assets

default: true

discover_lineage bool

Link charts and dashboards to the tables their SQL queries read

default: true

page_size int

Dashboards per search request

default: 100

Assets emitted

Metadata this plugin attaches to each discovered asset.

Dashboard

GrafanaDashboardFields

17 fields

The metadata fields the Grafana plugin emits for Dashboard assets.

uid string

Dashboard uid

id int

Dashboard numeric id

url string

Dashboard URL in Grafana

folder string

Title of the folder holding the dashboard

folder_uid string

Uid of the folder holding the dashboard

tags string[]

Tags set on the dashboard in Grafana

version int

Dashboard version number

schema_version int

Dashboard JSON schema version

created string

When the dashboard was created

updated string

When the dashboard was last saved

created_by string

User who created the dashboard

updated_by string

User who last saved the dashboard

provisioned bool

Whether the dashboard is managed by provisioning

refresh string

Auto refresh interval

time_from string

Start of the default time range

time_to string

End of the default time range

panel_count int

Number of panels, not counting rows and text panels

Chart

GrafanaChartFields

10 fields

The metadata fields emitted for Chart assets, one per dashboard panel.

panel_id int

Panel id within the dashboard

panel_type string

Grafana panel type, for example timeseries

chart_type string

Normalised chart type (Line, Table, Bar, ...)

dashboard string

Title of the dashboard holding the panel

dashboard_uid string

Uid of the dashboard holding the panel

datasource string

Name of the data source the panel queries

datasource_type string

Type of the data source the panel queries

datasource_uid string

Uid of the data source the panel queries

url string

URL of the panel in Grafana

target_count int

Number of queries the panel runs

Datasource

GrafanaDatasourceFields

9 fields

The metadata fields emitted for DataSource assets. Credentials are never read.

uid string

Data source uid

id int

Data source numeric id

type string

Data source plugin type, for example grafana-postgresql-datasource

type_name string

Data source plugin display name

url string

Address the data source connects to

database string

Database the data source connects to

is_default bool

Whether this is the default data source

read_only bool

Whether the data source is read-only in Grafana

access string

Access mode (proxy or direct)