Overview
The Error Tracking for Logs Explorer intelligently groups error logs into issues that help you understand and triage the root cause of errors by cutting through the noise and reducing thousands or millions of raw data points to a couple of issues consisting of diagnostic data such as stack traces, error distributions, and code snippets.
Datadog creates issues by computing a fingerprint for each error using some of its attributes such as the error type, the error message, or the stack trace. Errors with the same fingerprint are grouped together in the same issue. Each item listed in the Explorer is an issue and contains high-level information about the error:
- The error type and the error message
- The path to the file in which underlying errors are fired
- Important information about the issue’s lifetime:
- When it was first and last seen
- Graph of occurrences over time
- Number of occurrences in the selected time period
Time range
The time range appears on the top right of the Error Tracking Explorer as a timeline. This feature allows you to display issues having error occurrences within the selected time period. Change the time range by selecting a preset range from the dropdown.
Facets
Error Tracking automatically indexes a predefined list of attributes from your issues and creates facets for them. Clicking a facet filters the data to all available distinct values of the corresponding attribute, for the selected time period. It also provides some basic analytics, such as the number of issues represented. Facets allow you to pivot or filter your issues based on the given attribute.
Inspect an issue
Click on any issue to open the issue panel, which contains information about seasonality patterns, stack traces, and the error’s distribution across environments and sources. The What Happened section displays the first and last versions that are impacted by the error, plus an indication of when they were impacted. This metadata is kept beyond your standard log retention.
If the errors grouped into an issue have different stack traces, click Group Into Patterns.
Further Reading
Additional helpful documentation, links, and articles:
