This is documentation for the next version of Grafana. For the latest stable release, go to the latest version.
Tracing in Explore
Explore allows you to visualize traces from tracing data sources. This is available in Grafana v7.0+.
Supported data sources are:
For information on how to configure queries for the data sources listed above, refer to the documentation for specific data source.
- Header title: Shows the name of the root span and trace ID.
- Search: Highlights spans containing the searched text.
- Metadata: Various metadata about the trace.
Shows condensed view or the trace timeline. Drag your mouse over the minimap to zoom into smaller time range. Zooming will also update the main timeline, so it is easy to see shorter spans. Hovering over the minimap, when zoomed, will show Reset Selection button which resets the zoom.
Shows list of spans within the trace. Each span row consists of these components:
- Expand children button: Expands or collapses all the children spans of selected span.
- Service name: Name of the service logged the span.
- Operation name: Name of the operation that this span represents.
- Span duration bar: Visual representation of the operation duration within the trace.
Clicking anywhere on the span row shows span details.
- Operation name
- Span metadata
- Tags: Any tags associated with this span.
- Process metadata: Metadata about the process that logged this span.
- Logs: List of logs logged by this span and associated key values. In case of Zipkin logs section shows Zipkin annotations.
You can optionally expand the node graph for the displayed trace. Depending on the data source, this can show spans of the trace as nodes in the graph, or as some additional context like service map based on the current trace.
Trace to logs
Note: Available in Grafana 7.4 and later versions.
You can navigate from a span in a trace view directly to logs relevant for that span. This is available for Tempo, Jaeger, and Zipkin data sources at this moment. Refer to their relevant documentation for instructions on how to configure this feature.
Click the document icon to open a split view in Explore with the configured data source and query relevant logs for the span.
This visualization needs a specific shape of the data to be returned from the data source in order to correctly display it.
Data source needs to return data frame and set
frame.meta.preferredVisualisationType = 'trace'.
Data frame structure
|traceID||string||Identifier for the entire trace. There should be only one trace in the data frame.|
|spanID||string||Identifier for the current span. SpanIDs should be unique per trace.|
|parentSpanID||string||SpanID of the parent span to create child parent relationship in the trace view. Can be
|serviceName||string||Name of the service this span is part of.|
|serviceTags||TraceKeyValuePair||List of tags relevant for the service.|
|startTime||number||Start time of the span in millisecond epoch time.|
|duration||number||Duration of the span in milliseconds.|
|logs||TraceLog||List of logs associated with the current span.|
|tags||TraceKeyValuePair||List of tags associated with the current span.|
|warnings||string||List of warnings associated with the current span.|
|stackTraces||string||List of stack traces associated with the current span.|
|errorIconColor||string||Color of the error icon in case span is tagged with
Related Grafana resources
GrafanaCONline 2021 is coming June 7-17
Be the first to learn about exciting next-generation features in Grafana 8.0, be inspired by what community members are building, and attend expert-led sessions and workshops on Grafana, Prometheus, Loki logs, and more.
Unify your data with Grafana plugins: Splunk, MongoDB, Datadog, and more
Show how Grafana can be used to take data from multiple different sources and unify it, without disrupting the investments that are working today.
Demo: Getting started with Grafana Enterprise and observability
Join the Grafana Labs team for a 30-minute demo of how to get started with the Grafana Stack, so you can go from zero to observability in just a few minutes.