Search Traces

Epsagon traces contain valuable data about your application and how users are interacting with it, which can used to monitor you application health and to perform root-cause analysis for specific issues.

Trace Search is composed of three main sections:

  • Filters list - use this section to view only the traces you want to analyze.
  • Analytics - analyze success/error rates, average duration, and breakdown of error codes.
  • Results - all matching traces will appear in the results section. Scroll down to load more results.


Exploring filtered traces

Click on a trace row to view a map of all services called in its context and a timeline of how much time services took to respond.

Filter by Tags

Epsagon automatically tags traces to include the following:

  • application - every trace includes an application tag used to group services together
  • error - whether the trace call returned an error
  • exception_type - the name of the exception type raised by the trace request
  • resource_name - database name, endpoint path, etc...
  • resource_type - kafka, lambda, http, etc...
  • operation - POST, PutObject, etc...
  • duration - the duration of the event - can be filtered with larger or smaller than.
  • aws.* - tags relating to the AWS account and its services
  • azure.* - tags relating the azure account and its services
  • env.* - tags relating to the compute environment generating the trace, including Kubernetes tags
  • http.* - tags relating to HTTP requests

You can set additional tags to be indexed so that they'll be available for filtering. For example, you can choose to index user IDs, customer subscription plans, or the number of products in a shopping cart. Learn how to index tags..


Looking for possible values?

For any predefined filter, all possible values are already generated, so you can look for it while you type. You can put any other term even if it doesn't exist.

Sorting Search Results

By default, the results are sorted on a chronologically descending order, with newest events first.
Change the order of results on the top right of the filtering box:


Looking for the slowest performing events

Sorting by duration can help you find the slowest calls and events according to the filtering criteria.

Timeframe and Retention

To narrow down results, you can select a specific timeframe on the top left of the filtering box:

You can retain traces (and trace-based metrics) for longer periods than the 7-day default, contact sales for more information.

Did this page help you?