Maestro insights

Track flow health, flaky flows, and failure patterns for your Maestro end-to-end tests in the Maestro tab of EAS Insights.

Maestro insights is available on the Production and Enterprise plans. Learn more at EAS pricing.

The Maestro tab in the dashboard surfaces results from the Maestro end-to-end (E2E) tests you run in EAS Workflows with the maestro job. It helps you spot which flows fail or flake the most so you can keep your test suite healthy.

• To view insights, open your project in the EAS dashboard, select Insights from the navigation menu, and then select the Maestro tab.
• To start sending data, set up E2E tests by following Run E2E tests on EAS Workflows with Maestro. Results appear automatically as your tests run.

Insights are collected from the JUnit test report, which the maestro job produces by default (output_format: junit). If you set output_format to another value such as html, results aren't reported and won't appear here.

Each flow run falls into one of three states:

• Passed: Passed on the first attempt, with no retries.
• Flaky: Passed, but only after one or more retries.
• Failed: Did not pass.

Overview

The top of the tab summarizes activity for the selected time range, with each metric compared against the previous period of equal length:

• Maestro runs: How many flow runs completed.
• Pass rate: The share of runs that passed (a flaky run still counts as a pass).
• Flaky flows: How many distinct flows flaked at least once.
• Avg duration: Average run time.

Runs over time

A chart breaks runs down by day (or by hour or minute for shorter time ranges) into passed, flaky, and failed.

All Maestro flows

A table lists each flow with its Runs, Pass rate, Fails, Flake rate, P90 (90th percentile run duration), and Last run. Sort by any column to find the flows that need attention, and search by flow path.

Filters

Narrow the data down by workflow, status (passed, flaky, failed), tag, or branch.

Flow details

Select a flow to drill into its history, including its Maestro runs, pass rate, flaky runs, and P90 duration, plus:

• Error patterns: Failures grouped by error message, with a sample message and how often each pattern occurred.
• Recent runs: Individual runs with their status, duration, whether they flaked, and the workflow run, branch, and commit they came from.

More