grafana-wtf

About

grafana-wtf - grep through all Grafana entities in the spirit of git-wtf.

Attention!

This program can put significant load on your Grafana instance and the underlying database machinery. Handle with care!

Synopsis

Search Grafana API for string "weatherbase".

grafana-wtf find weatherbase

Display 50 most recent changes across all dashboards.

grafana-wtf log --number=50

Run with Docker:

# Access Grafana instance on localhost, without authentication.
docker run --rm -it --env GRAFANA_URL="http://host.docker.internal:3000" ghcr.io/panodata/grafana-wtf grafana-wtf info

# Access Grafana instance with authentication.
docker run --rm -it --env GRAFANA_URL="https://daq.grafana.org/grafana" --env GRAFANA_TOKEN="eyJrIjoiWHg...dGJpZCI6MX0=" ghcr.io/panodata/grafana-wtf grafana-wtf info

Screenshots

grafana-wtf find

grafana-wtf log

Setup

Install grafana-wtf

pip install grafana-wtf

Configure Grafana

Please take these steps to create an API key with your Grafana instance:

• Go to https://daq.example.org/grafana/org/apikeys.
• Choose "New API Key".
  • Key name: grafana-wtf
  • Role: Admin
• From the output curl -H "Authorization: Bearer eyJrIjoiWHg...dGJpZCI6MX0=" ..., please take note of the Bearer token. This is your Grafana API key.

Usage

Before running grafana-wtf, define URL and access token of your Grafana instance:

export GRAFANA_URL=https://daq.example.org/grafana/
export GRAFANA_TOKEN=eyJrIjoiWHg...dGJpZCI6MX0=

In order to ignore untrusted SSL certificates, append the ?verify=no query string to the GRAFANA_URL:

export GRAFANA_URL=https://daq.example.org/grafana/?verify=no

General information

# Display a bunch of meta information and statistics.
grafana-wtf info --format=yaml

# Display Grafana version.
grafana-wtf info --format=json | jq -r '.grafana.version'

Explore data sources

How to find unused data sources?

# Display all data sources and the dashboards using them, as well as unused data sources.
grafana-wtf explore datasources --format=yaml

# Display names of unused datasources as a flat list.
grafana-wtf explore datasources --format=json | jq -r '.unused[].datasource.name'

Explore dashboards

How to find dashboards which use non-existing data sources?

# Display some details of all dashboards, including names of missing data sources.
grafana-wtf explore dashboards --format=yaml

# Display only dashboards which have missing data sources, along with their names.
grafana-wtf explore dashboards --format=json | jq '.[] | select( .datasources_missing ) | .dashboard + {ds_missing: .datasources_missing[] | [.name]}'

Searching for strings

Find the string weatherbase throughout all dashboards and data sources:

grafana-wtf find weatherbase

Note

grafana-wtf will cache HTTP responses for 300 seconds by default. When running it with the --drop-cache option, it will drop its cache upfront.

Replacing strings

Replace all occurrences of ldi_v2 with ldi_v3 within dashboard with UID _JJ22OZZk:

grafana-wtf --select-dashboard=_JJ22OZZk replace ldi_v2 ldi_v3

In order to preview the changes, you should use the --dry-run option beforehand:

grafana-wtf --select-dashboard=_JJ22OZZk replace ldi_v2 ldi_v3 --dry-run

Display edit history

Watching out for recent editing activity on any dashboards?

# Display 50 most recent changes across all dashboards.
grafana-wtf log --number=50

Examples

For discovering more command line parameters and their arguments, please invoke grafana-wtf --help and have a look at grafana-wtf examples.

Development

git clone https://github.com/panodata/grafana-wtf
cd grafana-wtf

# Run all tests.
make test

# Run selected tests.
pytest --keepalive -vvv -k test_find_textual