Querying ROS data with Grafana

This guide explains how to query ROS data directly from ReductStore and visualize it in Grafana. It expands on the visualization workflow introduced in the ROS applications guide.

The example queries individual GPS messages from the public ROS dataset and plots interval-averaged latitude and longitude.

To run the preconfigured example, you need Docker with Docker Compose.

The Compose service uses the reduct/grafana:latest image, which includes Grafana and the ReductStore data source plugin. The public Play server provides the required ReductROS and ReductSelect extensions.

For a custom deployment, you need:

• Grafana 10 or newer;
• version 1.3.0 or newer of the ReductStore data source plugin; and
• a ReductStore server with the ReductROS and ReductSelect extensions.

Both extensions are available with ReductStore Pro.

See the Grafana integration guide for plugin installation and compatibility requirements.

Start Grafana

The ROS example includes the following Grafana service:

grafana:
  image: reduct/grafana:latest
  restart: unless-stopped
  ports:
    - "127.0.0.1:3000:3000"
  volumes:
    - ./grafana/datasources.yml:/etc/grafana/provisioning/datasources/datasources.yml:ro

It mounts this read-only provisioning file, which configures the ReductStore data source to query the public Play server:

apiVersion: 1
datasources:
  - name: ReductStore
    uid: reductstore
    type: reductstore-datasource
    jsonData:
      serverURL: https://play.reduct.store
    secureJsonData:
      serverToken: reductstore

Clone the documentation repository, enter the example directory, and start Grafana:

git clone https://github.com/reductstore/website.git
cd website/docs/ros/example
docker compose up -d grafana

Open http://localhost:3000 and sign in with the default username admin and password admin.

Demo credentials and data

Change the default Grafana credentials for any non-local deployment. The Play server and its reductstore token are shared demo resources; do not send private data to this server or use the demo token in production.

To query your own ReductStore instance, replace serverURL and serverToken in grafana/datasources.yml. That server must provide ReductROS and ReductSelect, and each raw ROS entry must have the $schema attachment needed for decoding.

Plot GPS coordinates

Create or edit a time-series panel, then configure its query:

1. Select the ReductStore data source.
2. Set the dashboard time range to Last 1 hour.
3. Select the orion bucket and the Pablo05/sensor/gps/fix entry.
4. Set Scope to Content Only.
5. Enable Combine frames.
6. Paste the following condition into the JSON editor and run the query:

{
  "#ext": [
    {
      "ros": {
        "extract": {}
      }
    },
    {
      "select": {
        "export": {
          "duration": "$__interval"
        },
        "sql": "SELECT latitude, longitude FROM ENTRY()"
      }
    },
    {
      "select": {
        "sql": "SELECT AVG(latitude) AS lat, AVG(longitude) as long FROM ENTRY()"
      }
    }
  ]
}

The #ext array runs an ordered server-side pipeline:

• ros.extract uses the entry's $schema attachment to decode each CDR message into JSON.
• The first select stage projects the latitude and longitude fields. Its export.duration setting batches rows according to their source timestamp span, using Grafana's resolved $__interval value.
• The second select stage averages each batch and emits the numeric lat and long series.

Grafana adjusts $__interval to the selected time range and panel width. This keeps the point density suitable for the panel while preserving more detail over narrower ranges. Combine frames keeps both output fields in one Grafana result frame.

The plugin flattens the combined JSON result, so the legend names may appear as $[0].lat and $[0].long rather than the bare SQL aliases.

Troubleshooting

If the panel is empty or one of the series is missing, verify that:

• the selected entry contains latitude and longitude fields;
• the entry has the $schema attachment required by ReductROS;
• ReductROS and ReductSelect are available on the queried server;
• the selected time range contains records for the entry; and
• Combine frames is enabled.