Selenium has a new trick up its sleeve to track failures
summary
Selenium wires OpenTelemetry tracing and structured event logs into the distributed Grid so every WebDriver command carries a trace id and spans showing router → session queue → distributor → node work; traces and embedded logs surface inputs, session ids, timestamps, 500 errors and stack traces. Demo uses Jaeger visualization to inspect end-to-end traces, reproduce a failure by removing geckodriver and see the exact span and exception. BrowserStack ships these telemetry feeds as a downloadable feature; metrics remain opt-in due to privacy and attribution concerns.
AI-written summary · about
chapters (17)
- 0:00Start
- 1:20Typical Observability Pipeline
- 2:04What Is Observability
- 7:17Why Do We Need Observability
- 7:46Selenium Grid
- 10:44Why We Need Observability in Selenium Grid
- 11:28Tracing
- 14:23How Do You Add the Context Using Attributes
- 16:26Demo
- 24:46Event Logs
- 28:40The Stack Trace
- 29:54Metrics
- 30:52Common Types of Matrix
- 33:06Why Are We Not Collecting Metrics in Selenium
- 37:17Observability
- 41:02Can We Add Jager Ui with Solenoid
- 42:21What Are some Practical Cases When Tracing Logs That Can Be Helpful as a User
