Get started with Distributed Traces

Distributed Traces provides you with a combination of analysis tools to gain insight into the transactions in your environment, combining code-level visibility, topology information, and metadata with the highest level of data granularity and fidelity. In this article, you will explore the Distributed Traces functionalities and how to leverage them to analyze requests in your environment.

Quick start

To get started with a distributed trace analysis

Go to Distributed Traces.
optional Use the Overview chart to understand the request count and response time over the selected timeframe.
The chart uses trace and request data, which has different data retention periods. For timeframes containing data older than 10 days, you can turn on the Show data retention toggle to better understand which data is available for which period directly from the chart.
optional In the table, expand the details of the trace you want to analyze for an overview of its topology, attributes, HTTP information, and timings.
Select the distributed trace name for in-depth trace analytics.
Select a time segment to expand the detail tabs.

Get an overview

To configure a view of the distributed traces in your environment

Go to Distributed Traces.
Configure a view by setting filters. To filter the distributed traces by
- Ingestion method
  - Select PurePaths to view PurePath® distributed traces captured by OneAgent.
  - Select Ingested traces to view distributed traces instrumented with other libraries.
- Service
  
  Go to Filter requests > Service name and enter the service name. Note that you can access this view also by going to Services > More (…) > Distributed traces for the service.

Export overview data

You can export the distributed traces overview table data in a comma-separated values (CSV) file.

Go to Distributed Traces.
optional Narrow down the table results by reconfiguring your view or via the search input field above the table.

The table lists up to the 3,000 most recent traces that are captured during the selected timeframe and within the selected management zone. Depending on the timeframe and the configured view, the list highlights the most important node for the trace.
To find a specific node of a trace
1. Go to Distributed Traces.
2. Filter the table by the trace ID.
3. Select the trace name.
4. In the Trace view, find the desired node.
In the lower-right corner of the page, select the Show export menu.

Select one of the following options.

Option	Exported data	Fields	Number of entries
Export visible data	The currently displayed area of the table, taking into account applied filters	Only visible fields	Up to 50 traces
Export table data	All table data, including traces that are not displayed on the current table page	All the available fields related to distributed traces	All 3,000 most recent traces

Analyze a trace execution

The distributed trace analysis view provides an end-to-end waterfall visualization of the requests in a trace.

To access the trace execution view

Go to Distributed Traces.
From the table, select the distributed trace name.
optional Go to the Execution breakdown to learn the trace time distribution.
Go to the waterfall chart to learn which services are traversed by the trace and in which order.
Select a request to visualize subsequent interactions.

Use the colors and positions of the horizontal bars in the chart to see which requests were executed synchronously (in parallel). They indicate both the sequence and response time of each of the requests.
Select a time segment to access in-depth trace analytics in detail tabs, down to code-level information.

Choose a detail tab to continue your analysis.

Availability	default
Content	Includes details on metadata, request headers, request parameters, and information about the proxy through which the request was sent.

Values obscured with five asterisks (*****) imply hidden confidential data, which can be unmasked by users with sufficient permissions. Three or fewer asterisks (***) are used for data aggregation purposes and can't be unmasked.

Availability	default
Content	Timing-specific details of services. Select View method hotspots for more details on timing contributions (for example, CPU, wait, sync, lock, and execution time).
Example	The example above shows timing details for the `/orange.jsf` web service call. In this case, you see that the request lasts 57.7ms.

Availability	depends on data input ¹ OneAgent
Content	Code execution of each and every request, including the code-level method executions and their timings, in the exact sequence of events. You can look at any request in the distributed traces view and navigate between the respective code-level trees. This gives you access to the full execution tree.
Example	In the example above, Dynatrace tells you exactly which method in the `/orange.jsf` request on the `easyTravel Customer Frontend` service called the respective web services and which specific web-service methods were called. The timings displayed here are the timings as experienced by the `easyTravel Customer Frontend`, which, in the case of calls to services on remote tiers, represent the client time.

The volume of code-level information varies depending on importance, timing, and estimated overhead. Typically, slower parts of a request contain more details.

Availability	depends on data input¹
Content	Log records in the full context of the transaction. All available span IDs are included by default. To filter by logs relevant to the selected waterfall node, turn on Strict span filter. To see all the logs for a trace, select View logs for this trace.
Example	Leverage log enrichment for traces to resolve problems

required Log enrichment.

Availability	depends on data input
Content	Occurring exceptions for failed requests.
Example	Drill-down to service failure causes

Availability

depends on service type OneAgent