The Pixie Datasource plugin allows you to visualize data from the Pixie observability platform in Grafana.
Instructions for installing the plugin can be found here.
To get started using the plugin, check out the tutorial.
The plugin requires a Pixie API key and cluster ID to execute queries. Grafana encrypts these values.
The plugin uses the Pixie Language (PxL) to query telemetry data collected by the Pixie platform.
If there are syntax errors in your PxL query, Grafana will display an error message.
If an output table has a column containing timestamps, the plugin will sort the table by the first timestamp column.
Grafana requires wide table format to visualize data in a Graph or Time series. If a query output table has a column of timestamps with the heading
time_, it will automatically convert the table to wide if necessary.
Grafana uses macros to add dashboard context to a query. The following macros are supported for PxL queries executed by plugin:
__time_fromwill be replaced by the start of the currently active time selection. Example usage:
df = px.DataFrame(table='http_events', start_time= __time_from)
__time_towill be replaced by the end of the currently active time selection. Example usage:
df = px.DataFrame(table='http_events', start_time= __time_from, end_time=__time_to)
__intervalwill be replaced by the suggested duration between time points. Interval is set by the Grafana UI in the Query options section in the Query tab. Example usage:
df.timestamp = px.bin(df.time_, __interval)