Company
Date Published
Author
Will McMullen
Word count
902
Language
English
Hacker News points
None

Summary

When a checkout flow breaks, customers disappear quickly. To set up observability for critical paths like customer checkout, Sentry can be used with minimal effort. Setting up monitoring and distributed tracing was straightforward, allowing the capture of performance metrics, errors, and tracing data across both Flask and React applications. A dashboard was created to monitor the most critical aspect of an eCommerce storefront: the checkout flow. The instrumentation in the Sentry SDK allowed for the attachment of attributes to spans, such as the number of items added to a cart, enabling the tracking of key signals in users' most critical journeys. This improved the ability to spot "outages" that aren't just uptime or spikes in errors. In this case, a complete dropoff in customers adding items to their carts was detected, and drilling into errors revealed issues with Flask's backend and React's frontend, which were resolved quickly. Sentry allows for the creation of alerts on span metrics, routing them out to the right team, providing clarity without context switching chaos.